JP2017184673A - Plant cultivation-supporting apparatus, plant cultivation-supporting system, method for supporting plant cultivation, program, and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a plant cultivation-supporting apparatus capable of accurately evaluating the state of a plant.SOLUTION: According to the present invention, there is provided a plant cultivation-supporting apparatus 10 that evaluates the state of a plant and supports cultivation of the plant. The plant cultivation-supporting apparatus 10 comprises: a first evaluation part that obtains predetermined data on the plant and derives an evaluation value of moisture stress of the plant based on the predetermined data; a second evaluation part that obtains image data of the plant and derives an evaluation value of increase or decrease in the number of leaves of the plant based on the image data; and a control part that determines a control to be applied to the plant based on the evaluation value of the moisture stress and the evaluation value of increase or decrease in the number of leaves and generates a control signal used for the control.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a plant cultivation support device, a plant cultivation support system, a plant cultivation support method, a program, and a storage medium.

  A technique called Speaking Plant Approach (SPA) that automatically diagnoses the state of plants and controls environmental conditions to a state suitable for plant growth has been proposed. This technology has been attracting attention in recent years as an effort to save labor in cultivation through recording of plant growth status and remote pathological diagnosis, and to realize high yield and high quality of agricultural products through moisture control. .

  Taking fruit vegetables as an example, if the effects of moisture control are shown as an example, if the fruit vegetables do not have sufficient water after fruiting, the fruit size may decrease due to water stress and the yield may decrease. On the other hand, it is also known that the fruit has a high sugar content by moderately applying water stress. Agricultural methods using this property produce high value-added crops through moisture control.

  The above-described farming method for producing high-sugar fruits can increase income per unit area due to high added value of crops, but has the following problems. The first problem is that the timing and amount of irrigation depends on the experience and intuition of the operator, and trial and error and skill are required before appropriate control can be performed. The second issue is that the optimal amount and timing of irrigation changes depending on the weather and the degree of crop growth. If this is mistaken, the crop will die due to excessive water stress, and the fruit quality will deteriorate due to insufficient water stress. It is to connect.

  In view of these circumstances, the purpose of realizing a device that performs liquid supply control for plant cultivation is to determine the timing of liquid supply based on the projected area of the plant leaf obtained using the image input device. Techniques to do this are disclosed. More specifically, the liquid supply is controlled by issuing a liquid supply signal when the ratio between the reference value of the projected area and the current projected area (projected area ratio) falls below a threshold value. The reference value of the projected area described above is continuously updated with the maximum value of the most recent projected area to follow the growth of the plant, and when the liquid is supplied, it is overwritten with the projected area at that time. It adapts to the temporary reduction of the projected area due to shaving.

  However, the method for updating the reference value of the projected area in the above technique may not accurately evaluate the state of the plant for the following reason.

  First, it cannot cope with the movement of stocks by oblique or hanging attraction. In other words, in the long-stage cultivation of vinyl houses and cherry tomatoes with a low eave height, oblique and hanging invitations that lead the strain in an oblique direction according to the growth are widely performed for the purpose of suppressing the height of the strain. .

  Unlike an upright invitation that raises the stock upright, in the oblique or hanging attraction, the upper position of the stock is moved horizontally according to the growth, so the relevant part of the stock of interest is within the observation range of the image input device. In some cases, it may deviate, or an adjacent strain may newly enter from outside the observation range.

  In the latter case in particular, the above method may cause a deviation between the projected area ratio and the actual moisture stress. Specifically, the projected area increases regardless of moisture stress, the projected area ratio increases, or the projected area reference value is updated to reset the projected area ratio to 1. Stress can be underestimated.

  Second, if the liquid supply is not sufficient, the reference value is estimated to be unreasonably low. That is, if the amount of liquid supply is not sufficient with respect to the water stress, the water stress may not be completely eliminated. In the above technology, the reference value is temporarily replaced with the current value at the time of liquid supply, and then sequentially updated with the maximum value.However, under the above conditions, the reference value does not return to an appropriate level, and the subsequent water stress is unduly low. There is a risk of being evaluated.

  This invention is made | formed in view of said point, and it aims at providing the plant cultivation assistance apparatus which can evaluate the state of a plant accurately.

  The plant cultivation support device is a plant cultivation support device that evaluates the state of a plant and supports the cultivation of the plant, obtains predetermined data regarding the plant, and determines water stress of the plant based on the predetermined data. A first evaluation unit for deriving an evaluation value, a second evaluation unit for obtaining image data of the plant, and deriving an evaluation value of increase / decrease in leaf of the plant based on the image data, and an evaluation value of the water stress And a control unit that determines a control to be applied to the plant based on the evaluation value of the increase / decrease of the leaves and generates a control signal used for the control.

  According to the disclosed technology, it is possible to provide a plant cultivation support device that can accurately evaluate the state of a plant.

It is a figure explaining the plant cultivation assistance system which concerns on 1st Embodiment. It is a figure which illustrates the function of CPU which comprises the plant cultivation assistance apparatus which concerns on 1st Embodiment. It is an example of the flowchart which shows the process of the plant cultivation assistance system which concerns on 1st Embodiment. It is a figure explaining the plant cultivation assistance system which concerns on 4th Embodiment. It is an example of the flowchart which shows the process of the plant cultivation assistance system which concerns on 4th Embodiment.

  Hereinafter, embodiments will be described with reference to the drawings. In addition, in each drawing, the same code | symbol is attached | subjected to the same component and the overlapping description may be abbreviate | omitted. Further, in this specification, water and fertilizer may be collectively referred to as “nutrient solution”, and the act of supplying a nutrient solution to a plant may be referred to as “liquid supply”.

<First Embodiment>
FIG. 1 is a diagram illustrating a plant cultivation support system according to the first embodiment. Referring to FIG. 1, a plant cultivation support system 1 is a system that supports plant cultivation by evaluating the state of a plant, and includes a plant cultivation support device 10 and a camera 20 as main components. ing. However, in FIG. 1, the plant cultivation support device 10 and the camera 20 are illustrated as separate bodies, but both may be integrated into one housing. The plant state includes, for example, leaf wilt, moisture stress, and growth.

  The plant cultivation support apparatus 10 includes an interface 11, a CPU (Central Processing Unit) 12, and a storage unit 13. The interface 11, CPU 12, and storage unit 13 are connected to each other via a bus 14. The plant cultivation support device 10 is, for example, a PC (personal computer).

  The interface 11 has a function of communicating with the camera 20. The CPU 12 has a function of comprehensively controlling various operations in the plant cultivation support system 1. The storage unit 13 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The storage unit 13 may include a nonvolatile memory such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

  For example, the CPU 12 executes the program stored in the ROM in cooperation with the RAM, thereby realizing, for example, the functions of the first evaluation unit 121, the second evaluation unit 122, and the control unit 123 illustrated in FIG. The In addition, FIG. 2 is a figure which illustrates the function of CPU which comprises the plant cultivation assistance apparatus which concerns on 1st Embodiment. The CPU 12 may have functions other than those shown in FIG. 2 as necessary.

  The camera 20 is configured to be able to communicate with the interface 11 of the plant cultivation support apparatus 10. Communication may be wired or wireless. For example, the camera 20 can acquire image data of the plant 500 as a subject in response to a predetermined time interval or an instruction from the plant cultivation support device 10. As the camera 20, for example, a CCD (Charge Coupled Device) camera, a CMOS (Complementary Metal Oxide Semiconductor) camera, or the like can be used.

  The image data acquired by the camera 20 is transmitted to the plant cultivation support apparatus 10 and stored in the storage unit 13 via the interface 11. The stored image data is read from the storage unit 13 by the CPU 12. The first evaluation unit 121 acquires image data, derives an evaluation value of the water stress of the plant 500 based on the acquired image data, and estimates the water stress state. The evaluation value of water stress is an evaluation value representing the degree of water shortage in plants.

  The evaluation value of water stress can be derived based on changes in the appearance of plants. For example, taking a change in the projected area of a leaf as an example of a change in the appearance of a plant, the cause of the decrease in the projected area of the leaf is leaf scraping that removes unnecessary leaves in addition to the leaf wilting accompanying the worsening of water stress. Examples include cases where the stocks that were captured by the camera were moved out of the angle of view when the attracting string was moved by hanging invitations.

  On the other hand, the factor that increases the projected area of the leaves is not only the recovery from leaf wilting caused by the relaxation of moisture stress, but also the new stock is reflected in the angle of view of the camera when the attracting string is moved by hanging attraction. (Hereinafter, the movement of the attracting string and the change of the fixed position of the plant in the attracting string are collectively referred to as an attracting work).

  When a change in the appearance of a plant (for example, a change in the projected area of a leaf) occurs due to a factor that is not moisture stress, the following problem may occur due to misjudging it as caused by moisture stress. For example, if water stress is overestimated, excessive water supply may result in a decrease in fruit quality such as a decrease in sugar content or a disease such as root rot. On the other hand, if the water stress is underestimated, there is a risk that the quality of the fruit, such as hardening of the skin, will be reduced or the plant will die if not given enough water.

  Therefore, in the plant cultivation support apparatus 10, the second evaluation unit 122 obtains image data of the plant 500 in order to accurately identify whether the cause of the appearance change of the plant 500 is water stress or other factors. Then, an evaluation value of increase / decrease of the leaves of the plant 500 is derived based on the obtained image data. The 2nd evaluation part 122 identifies whether the cause of the external appearance change of the plant 500 is a water stress or another factor based on the derived evaluation value of the increase or decrease of the leaf. Here, as an example of other factors, management work such as leaf scraping work or attracting work can be cited.

  The controller 123 should be applied to the plant 500 based on the water stress evaluation value derived from the image data by the first evaluation unit 121 and the leaf increase / decrease evaluation value derived from the image data by the second evaluation unit 122. Control is determined and a control signal used for control is generated. For example, the control unit 123 determines whether or not the liquid supply to the plant 500 is necessary, determines the liquid supply as a control to be applied to the plant 500, and generates a control signal used for the liquid supply.

  In addition to the evaluation value of the water stress derived by the first evaluation unit 121, the control unit 123 uses the evaluation value of the increase / decrease of the leaf derived by the second evaluation unit 122 to allow the control unit 123 to perform control (for example, , Liquid supply) can be determined with high accuracy.

  The plant cultivation support device 10 is configured to be connectable to the notification unit 70. When the control unit 123 determines that liquid supply is necessary, a control signal generated by the control unit 123 is transmitted to the notification unit 70, and the notification unit 70 can notify that liquid supply is necessary.

  The notification that the liquid supply is necessary can be realized by, for example, a method of sounding a buzzer or a method of flashing a rotating lamp. The notification that the liquid supply is necessary may be realized by a method of contacting a preset contact address by e-mail or telephone. Alternatively, the notification may be made visually using a smartphone dedicated application instead of email or telephone.

  The plant cultivation support device 10 is configured to be connectable to the storage medium mounting unit 80. The storage medium mounting unit 80 can detachably mount a storage medium storing a program executed by the CPU 12. Examples of storage media that can be mounted on the storage medium mounting unit 80 include USB (Universal Serial Bus) memory, SD memory, CD (Compact Disc), DVD (Digital Versatile Disc), BD (Blu-ray (registered trademark) Disc), and the like. It is. However, the program executed by the CPU 12 may be downloaded to the storage unit 13 via the network without using the storage medium mounting unit 80.

  The camera 20 may be installed anywhere as long as the plant 500 can be caught within the viewing angle, but it is desirable to install the camera 20 at a position near the plant 500 in the vertical direction.

  Next, a plant cultivation support method using the plant cultivation support system 1 will be described. FIG. 3 is an example of a flowchart showing processing of the plant cultivation support system according to the first embodiment. The process of the plant cultivation support system 1 will be described in detail with reference to FIG.

  First, in step S <b> 101, the first evaluation unit 121 of the CPU 12 acquires image data of the plant 500 from the camera 20 through the interface 11 at a predetermined timing. The predetermined timing is, for example, the timing at which the camera 20 receives a trigger signal from the CPU 12 via the interface 11. Alternatively, the timing may be determined based on a preset time or time interval.

  Next, in step S102, the plant cultivation support apparatus 10 estimates a moisture stress state based on the image data acquired in step S101. Specifically, the first evaluation unit 121 of the CPU 12 analyzes the image data obtained from the camera 20, estimates the water stress state of the plant 500 based on the image data, and derives the evaluation value of the water stress. Examples of the method for estimating the water stress state from the image data include the following.

  The first example is estimation based on the projected area of the leaves. When plants are exposed to water stress, the projected area of the leaves decreases due to the wilting of the leaves. Therefore, it is possible to estimate the projected area of the leaf in a state where no moisture stress is applied in advance as a reference value, and to estimate the degree of moisture stress based on the ratio between the projected area at the current time and the reference value. For example, the projected area of the leaf of the plant can be obtained from the image data, and the degree of moisture stress can be estimated based on the ratio between the maximum value of the most recent projected area and the current projected area.

  The second example is an estimation based on the leaf shape. When the leaves wither due to water stress, the appearance of the leaves in the image data changes. Therefore, it is possible to estimate moisture stress by quantifying the shape change. For example, by extracting the leaves from the image data and comparing the ratio of the length and width with a preset value, or by learning the image data of the leaves according to the degree of wilting and pattern matching, A method of identifying the degree of time wilting can be mentioned.

  The third example is an estimation based on a representative value of leaves. Leaves usually sag when they wither. Therefore, for example, a plant can be photographed from the horizontal direction to obtain a representative value such as the center of gravity of the leaf, and moisture stress can be estimated based on the amount of change from the reference position before wilt.

  Hereinafter, unless otherwise specified, the description will be made assuming that the projected area of the leaf is used when estimating the moisture stress state from the image data, but the present invention is not limited to this.

  Next, in step S103, the plant cultivation support apparatus 10 derives an evaluation value for increase / decrease in the leaves of the plant 500 based on the image data acquired in step S101. Specifically, the second evaluation unit 122 of the CPU 12 analyzes the image data obtained from the camera 20, extracts a leaf region from the image data, derives a representative value of the leaf region, and determines a predetermined representative value. The evaluation value of the increase or decrease of the leaf is derived based on the amount of change in the time interval. In this way, by analyzing the change over time in the representative value of the leaf region, it is possible to detect the fluctuation in the plant by detecting the increase or decrease of the leaf.

  In this embodiment, an example in which the area (projected area) of a leaf region is used as a representative value of the leaf region is shown. Specifically, the amount of change per unit time of the projected area of the leaves shown in the image data, that is, the first derivative of the time series data of the projected area is used. The progress and recovery of leaf wilt caused by water stress is a physiologic response of the plant and progresses slowly. On the other hand, since the management work is usually performed in a short time, the appearance change accompanying it rapidly proceeds. By taking the first derivative of the time-series data of the projected area, it is possible to identify both using the difference in time response.

  The first derivative of the time series data of the projected area is calculated by subtracting the projected area at the second time, which is a predetermined time interval from the first time, from the projected area at the first time. If the value of the obtained first derivative is an absolute value and is larger than a preset threshold value, it can be determined that the number of leaves has increased or decreased because of the management work.

  A positive threshold value and a negative threshold value may be provided without taking the absolute value of the value of the first derivative. Specifically, when the value of the first derivative is negative, if the value is smaller than the negative threshold, it is determined that the increase or decrease of the leaf has occurred due to management work, and the value of the first derivative is positive. If it is larger than the positive threshold, it may be determined that the increase or decrease of the leaf has occurred because the management work has occurred.

  The first derivative of the projected area becomes negative in relation to water stress when the wilting progresses, and conversely becomes positive when the wilting recovers. Since the former usually proceeds more slowly than the latter, the above-described negative threshold value may be set smaller in absolute value than the positive threshold value.

  The determination of the occurrence of management work by deriving the evaluation value of the increase / decrease in leaves may be performed sequentially when the plant cultivation support apparatus 10 receives image data from the camera 20, or the water stress state is determined in step S102. You may implement only when predetermined conditions are satisfy | filled.

  Next, in step S104, the plant cultivation support device 10 determines whether or not liquid supply is necessary based on the evaluation value of the water stress and the evaluation value of the increase / decrease of the leaves. Specifically, the control unit 123 of the CPU 12 determines whether or not the liquid supply to the plant 500 is necessary, the water stress evaluation value derived by the first evaluation unit 121 in step S102, and the second evaluation unit 122 in step S103. Judgment is made based on the evaluation value of the increase or decrease of the leaves derived from. The necessity of liquid supply is, for example, that the evaluation value of water stress exceeds a preset threshold value, and the evaluation value of leaf increase / decrease exceeds a preset threshold value (no management operation has been detected) ) Can be determined based on that.

  It should be noted that the water stress evaluation value may be updated based on the leaf increase / decrease evaluation value and then compared with the threshold value. Specifically, when it is determined that the management work has occurred based on the evaluation value of the increase or decrease of the leaf, the evaluation value of the water stress before the management work occurs or the water stress A method of overwriting with the evaluation value of the moisture stress in the absence state is conceivable.

  If it is determined in step S104 that liquid supply is necessary (in the case of Yes), the process proceeds to step S105. If it is determined that liquid supply is not required (in the case of No), the process proceeds to step S106.

  In step S105, the notification unit 70 performs a predetermined notification. That is, when the control unit 123 determines that liquid supply is necessary, the CPU 12 transmits information to the notification unit 70 and notifies the notification unit 70 that liquid supply is necessary. A specific example of the notification that liquid supply is necessary is as described above.

  The above steps S101 to S105 are a series of processes performed in association with one acquisition of image data. When these steps are completed, it is determined in step S106 whether or not to end the control. If it is determined in step S106 that the process is to be ended (in the case of Yes), the control is ended. If it is determined that the process is not to be ended (in the case of No), the process returns to step S101. The processing of S101 to S105 is repeated.

  In step S103, an example in which the presence / absence of management work is estimated based on the evaluation value of increase / decrease in the amount of leaves is shown. However, the management work is only an example of the cause of increase / decrease in the amount of leaves. As described above, an increase or decrease in the amount of leaves leads to a misjudgment of the water stress state, so capturing the increase or decrease in the amount of leaves is the essence of the present invention, and estimating the occurrence of management work is the present invention. Is not the essence of

  In step S105, when it is determined that liquid supply is necessary, the contact information registered in advance is notified, but the present invention is not limited to this. For example, liquid supply means (not shown) may be used to automatically supply liquid to a plant by a preset amount according to the moisture stress state.

  As described above, in the plant cultivation support device according to the first embodiment, when estimating the water stress state of the plant based on the change in the appearance of the plant, the water stress that is easily confused with the change in the appearance of the plant accompanying the water stress. Changes in the appearance of the plant due to factors different from those (for example, management work such as leaf scraping and attraction) are detected.

  More specifically, the water stress evaluation value is derived using an arbitrary method including the existing technology, and the leaf increase / decrease is detected based on the image data to derive the leaf increase / decrease evaluation value. Then, it is determined whether or not management work has occurred based on the evaluation value of the increase / decrease of the leaves. By integrating these results, if there is a possibility that an erroneous estimation resulting from the management work has occurred in the estimation result of the moisture stress state, an incorrect liquid supply determination can be avoided by suppressing it.

  That is, the plant to be observed is constantly monitored, and when determining the necessity of liquid supply based on the evaluation value of water stress, it is detected whether or not management work has occurred based on the evaluation value of leaf increase / decrease. By adding to the determination material, it is possible to suppress an erroneous determination of whether or not a liquid supply is necessary based on a change in the appearance of the plant accompanying a management operation or the like, and it is possible to realize a plant cultivation support system with improved estimation accuracy of water stress. By utilizing the evaluation result of water stress, the amount and timing of irrigation can be controlled appropriately, and high-yield and high-quality crops can be produced stably and easily.

  The plant cultivation support system 1 may be configured to use a storage device in which image data acquired in advance is used instead of the camera 20 and read the image data from the storage device when the plant cultivation support device 10 is necessary. The camera 20 and the storage device are an example of an image data providing unit that provides image data to the plant cultivation support device 10.

<Modification of First Embodiment>
In the modified example of the first embodiment, an example of a plant cultivation support apparatus that derives an evaluation value of increase / decrease in the amount of leaves by a variation in projected area for each partial region is shown. In the modification of the first embodiment, the description of the same components as those of the already described embodiments may be omitted.

  In the first embodiment, the presence / absence of management work is determined based on the evaluation value of increase / decrease in leaves without limiting the area in the image data, but the present invention is not limited to this. Since the management work is normally performed independently for each stock, there may be a case where a management work is performed on a plurality of stocks included in the image data and a case where the management work is not performed.

  At this time, there is a possibility that the increase / decrease of the leaf accompanying the management work may occur only locally, and if the entire image data is evaluated, it may not be detected as a large variation. Therefore, in the modification of the first embodiment, the image data is divided into a plurality of partial areas, and the value of the first derivative of the projected area is evaluated for each partial area to derive the evaluation value of the increase / decrease in the leaves. Determine whether there is management work.

  Specifically, first, the image data is divided into partial areas based on preset criteria. As a dividing method, a method of dividing at equal intervals and a method of dividing so that the projected areas of the leaves are approximately the same can be considered. In addition, it is desirable to ignore a partial region having an extremely small amount of leaves because it has low noise resistance.

  Next, the first derivative of the projected area is derived for each partial region, and management work is determined based on the evaluation value of the increase / decrease of the leaf for each local region as in step S103 of FIG. Finally, the determination results for each local area are integrated to determine the presence or absence of management work. As a method of integrating the determination results for each local area, when a predetermined number of local areas are determined to have management work, or when a predetermined number of spatially continuous local areas are determined to have management work In addition, there is a method for determining that there is a management work at the time.

  According to the plant cultivation support device according to the modification of the first embodiment, the management work is determined based on the evaluation value of the increase or decrease of the leaves for each local region, and the result is integrated to determine the presence or absence of the management work. As a result, the detection sensitivity can be improved as compared with the case where the entire image data is determined collectively. Other effects are the same as those in the first embodiment.

<Second Embodiment>
In 2nd Embodiment, the example of the plant cultivation assistance apparatus which derives | leads-out the evaluation value of the increase / decrease in the amount of leaves by the gravity center position fluctuation | variation of the whole image is shown. In the second embodiment, description of the same components as those of the already described embodiments may be omitted.

  Since the configuration and processing flow of the plant cultivation support apparatus according to the second embodiment are the same as those in FIGS. However, the processing content in step S103 of FIG. 3 is different from the plant cultivation support apparatus according to the first embodiment.

  In the first embodiment, the evaluation value for the increase / decrease in the amount of leaves is determined based on the temporal change in the projected area. In the second embodiment, the evaluation value is determined based on the temporal change in the barycentric position of the projection area. . In other words, in the first embodiment, the area of the leaf area is used as the representative value of the leaf area. In the second embodiment, the barycentric position of the leaf area is used as the representative value of the leaf area. Use. In the case of the projected area, if the increase and decrease of the area, such as translation of the plant, are about the same, it may not be considered as a difference.

  In step S103 of FIG. 3, the second evaluation unit 122 of the CPU 12 analyzes the image data obtained from the camera 20, and derives an evaluation value of the increase / decrease in the leaf based on the increase / decrease in the amount of the leaf reflected in the image data.

  Specifically, the amount of change per unit time of the centroid position of the projection area obtained by extracting only the leaf area in the image data, that is, the first derivative of the time series data of the centroid position of the projection area is used. The progression and recovery of leaf wilting due to water stress usually occurs simultaneously in all the strains present within the observation range, so there is no spatial bias in the variation.

  On the other hand, the increase or decrease of the plant leaf accompanying the management work occurs locally, so there is a spatial bias in the variation. Since the barycentric coordinates are affected by the spatial deviation of the fluctuation, this is used in the present embodiment. The first derivative of the barycentric coordinates is calculated by subtracting the barycentric coordinates at the second time, which is a predetermined time interval from the first time, from the barycentric coordinates at the first time.

  The obtained first derivative value is an absolute value, and if it is larger than a preset threshold value, it is determined that a management operation has occurred. The x and y coordinates of the barycentric coordinates are handled independently, and it is determined that a management operation has been detected when at least one or both of the first derivative of the x coordinate and the first derivative of the y coordinate satisfy the above-described conditions. May be.

  Further, the square average of the first derivative of the x coordinate and the first derivative of the y coordinate may be taken, and the presence / absence of management work may be determined by comparing the value with a threshold value.

  The determination of the presence or absence of management work by deriving the evaluation value of the increase / decrease in leaves may be performed sequentially when the plant cultivation support apparatus 10 receives image data from the camera 20, or the water stress state is predetermined in step S102. It may be performed only when the above condition is satisfied.

  According to the plant cultivation support device according to the second embodiment, whether or not a management operation has occurred by deriving an evaluation value of the increase or decrease of the leaf based on the position change of the barycentric coordinates of the projection region in the plant to be observed. This makes it possible to suppress erroneous determination of whether or not liquid supply is necessary. Even if the projected area does not vary greatly due to the management work, it is possible to detect whether or not the management work has occurred from the variation in the spatial deviation. Other effects are the same as those in the first embodiment.

<Modification of Second Embodiment>
In the modification of 2nd Embodiment, the example of the plant cultivation assistance apparatus which derives | leads-out the evaluation value of increase / decrease in the quantity of a leaf by the gravity center position fluctuation | variation for every partial region is shown. In the modification of the second embodiment, the description of the same components as those of the already described embodiments may be omitted.

  In the second embodiment, the presence / absence of management work is determined without limiting the area in the image data, but the present invention is not limited to this. As described above, the change in the appearance of the plant accompanying the management work occurs locally, and therefore can be detected as a change in the position of the barycentric coordinates.

  On the other hand, if the locality of the appearance change is taken into account, it is expected that the detection sensitivity of the change is improved by evaluating the position change of the barycentric coordinates in a narrower range. Therefore, in the modification of the second embodiment, the image data is divided into a plurality of partial areas, and the value of the first derivative of the center of gravity position of the projection area is evaluated for each partial area to derive the evaluation value of the increase or decrease of the leaves. To determine whether there is any management work.

  Specifically, first, the image data is divided into partial areas based on preset criteria. The dividing method is the same as that of the modified example of the first embodiment. Next, the first derivative of the center-of-gravity coordinates of the projection area is derived for each partial area, and the management work is determined based on the evaluation value of the increase / decrease of the leaves for each local area, as in the modification of the first embodiment.

  Finally, the determination results for each local area are integrated to determine the presence or absence of management work. The method for integrating the determination results for each local area is the same as that of the modification of the first embodiment.

  According to the plant cultivation support device according to the modification of the second embodiment, the management work is determined based on the evaluation value of the increase / decrease in the leaves for each local region, and the result is integrated to determine the presence / absence of the management work. As a result, the detection sensitivity can be improved as compared with the case where the entire image data is determined collectively. Other effects are the same as those in the first embodiment.

<Third Embodiment>
In 3rd Embodiment, the example of the plant cultivation assistance apparatus which derives | leads-out the evaluation value of the increase / decrease in the amount of leaves by the magnitude | size of the increase part of the projection area | region of the whole image, or the reduction | decrease part is shown. Note that in the third embodiment, description of the same components as those of the already described embodiments may be omitted.

  Since the configuration and processing flow of the plant cultivation support apparatus according to the third embodiment are the same as those shown in FIGS. However, the processing content in step S103 of FIG. 3 is different from the plant cultivation support apparatus according to the first embodiment.

  The difference between the projected areas at the two times is divided into a portion where the projected area increases and a portion where the projected area decreases. In the approach of the first embodiment using the first derivative of the projected area, these are added together and evaluated, so that it is not possible to detect fluctuations under conditions where both exist in the same size. In the third embodiment, the detection sensitivity of fluctuation is increased by evaluating these individually.

  In step S103 of FIG. 3, the second evaluation unit 122 of the CPU 12 analyzes the image data obtained from the camera 20, and derives an evaluation value of the increase / decrease in the leaf based on the increase / decrease in the amount of the leaf reflected in the image data.

  First, a difference image is created by subtracting a projection area at a second time that is back from the projection area at the first time by a predetermined time interval from the first time. Next, the number of pixels (increasing pixels) that take a positive value in the difference image (the number of increasing pixels) and the number of pixels (decreasing pixels) that take a negative value (the number of decreasing pixels) are counted.

  The increase pixel number and the decrease pixel number are each compared with a predetermined threshold value, and it is determined that the management work has occurred when at least one is greater than the threshold value or both are greater than the threshold value. Note that the increased pixel number and the decreased pixel number may be limited to only one of them.

  The determination of the occurrence of management work by deriving the evaluation value of the increase / decrease in leaves may be performed sequentially when the plant cultivation support apparatus 10 receives image data from the camera 20, or the water stress state is determined in step S102. You may implement only when predetermined conditions are satisfy | filled.

  According to the plant cultivation support device according to the third embodiment, whether or not the management work has occurred is detected based on the size of the increase or decrease of the projection area in the plant to be observed. By using it as a determination material when determining the necessity, it is possible to suppress erroneous determination of the necessity of liquid supply. Even if the projected area does not vary greatly due to the management work, it is possible to detect whether or not the management work has occurred from the variation in the spatial deviation. Other effects are the same as those in the first embodiment.

<Variation 1 of the third embodiment>
In the first modification of the third embodiment, an example of a plant cultivation support device that derives an evaluation value of increase / decrease in the amount of leaves based on the size of the increase or decrease of the projection region for each partial region is shown. In the first modification of the third embodiment, the description of the same components as those of the already described embodiments may be omitted.

  In the third embodiment, the presence / absence of management work is determined without limiting the area in the image data, but the present invention is not limited to this. As described above, since the appearance change of the plant accompanying the management work occurs locally, it can be expected that the detection sensitivity of fluctuation is improved by evaluating the increase or decrease of the projection area in a narrower range. . Therefore, in the first modification of the third embodiment, the image data is divided into a plurality of partial areas, and the presence / absence of management work is determined by evaluating the number of increased or decreased pixels for each partial area.

  First, the image data is divided into partial areas based on preset criteria. The dividing method is the same as that of the modified example of the first embodiment. Next, the number of increased or decreased pixels in the projection area is derived for each partial area, and management work is determined based on the evaluation value of the increase / decrease in leaves for each local area, as in step S103 of FIG. Finally, the determination results for each local area are integrated to determine the presence or absence of management work. The method for integrating the determination results for each local area is the same as that of the modification of the first embodiment.

  According to the plant cultivation support device according to the first modification of the third embodiment, the management work is determined based on the evaluation value of the increase or decrease of the leaves for each local region, and the result is integrated to determine whether or not there is the management work. By making the determination, the detection sensitivity can be improved as compared with the case where the entire image data is determined collectively. Other effects are the same as those in the first embodiment.

<Modification 2 of the third embodiment>
In a second modification of the third embodiment, an example of a plant cultivation support apparatus that derives an evaluation value for an increase or decrease in the amount of leaves by a change in the center of gravity position of an increase or decrease in the projection area is shown. In the second modification of the third embodiment, the description of the same components as those of the already described embodiments may be omitted.

  In the third embodiment, the management work is detected based on the number of increased pixels and decreased pixels, but the present invention is not limited to this. The locality in the appearance change of the plant accompanying the management work means that there is a spatial bias in the appearance positions of the increase pixel and the decrease pixel that are the differences between the projection areas separated by a predetermined time.

  This spatial deviation can be detected as a change with time in the barycentric coordinates as in the second embodiment. That is, every time image data is acquired, the center of gravity position of increasing pixels and decreasing pixels is obtained and accumulated as time-series data, and the occurrence of management work is determined by comparing the absolute value of the first derivative of the data with a threshold value. Judgment can be made. Note that, similarly to the first modification of the third embodiment, the determination may be performed after performing the determination for each partial region. Other effects are the same as those in the first embodiment.

<Fourth embodiment>
In 4th Embodiment, the example of the plant cultivation assistance system provided with the distance sensor is shown. Note that in the fourth embodiment, descriptions of the same components as those of the above-described embodiments may be omitted.

  FIG. 4 is a diagram illustrating a plant cultivation support system according to the fourth embodiment. Referring to FIG. 4, the plant cultivation support system 2 is different from the plant cultivation support system 1 (see FIG. 1) in that a distance sensor 30 is added.

  The distance sensor 30 is configured to be able to communicate with the interface 11 of the plant cultivation support device 10. Communication may be wired or wireless. The distance sensor 30 acquires distance data to the subject plant 500 in response to, for example, a predetermined time interval or an instruction from the plant cultivation support device 10.

  As the distance sensor 30, for example, a sensor (for example, a monocular camera, a stereo camera, or the like) that employs a passive method such as a lens focus method or a passive stereo method can be used. Further, as the distance sensor 30, for example, a sensor (for example, an optical radar, a projector camera system, or the like) that employs an active type such as a TOF (Time-of-Flight) method or an active stereo method may be used. The distance sensor 30 may be composed of a plurality of sensors of the same type or a plurality of types of sensors.

  The distance data acquired by the distance sensor 30 is provided to the plant cultivation support device 10 and stored in the storage unit 13 via the interface 11. The stored distance data is read from the storage unit 13 by the CPU 12. The first evaluation unit 121 obtains distance data, derives an evaluation value of the water stress of the plant 500 based on the obtained distance data, and estimates a water stress state. By evaluating the water stress using the distance data, it is possible to capture leaf fluctuations that cannot be captured by image data.

  Similar to the first embodiment, the image data acquired by the camera 20 is transmitted to the plant cultivation support apparatus 10 and stored in the storage unit 13 via the interface 11. The stored image data is read from the storage unit 13 by the CPU 12. The second evaluation unit 122 obtains image data, and derives an evaluation value of increase / decrease in the leaves of the plant 500 based on the obtained image data. The 2nd evaluation part 122 identifies whether the cause of the external appearance change of the plant 500 is a water stress or another factor based on the derived evaluation value of the increase or decrease of the leaf. For example, an evaluation value of leaf increase / decrease is derived to estimate whether or not management work has occurred.

  In addition, the control unit 123 applies the plant 500 based on the water stress evaluation value derived from the distance data by the first evaluation unit 121 and the leaf increase / decrease evaluation value derived from the image data by the second evaluation unit 122. The control to be performed is determined, and a control signal used for the control is generated. For example, the control unit 123 determines whether or not the liquid supply to the plant 500 is necessary, determines the liquid supply as a control to be applied to the plant 500, and generates a control signal used for the liquid supply.

  As in the first embodiment, when the control unit 123 determines that liquid supply is necessary, the control signal generated by the control unit 123 is transmitted to the notification unit 70 and the notification unit 70 needs liquid supply. Notification can be made.

  The camera 20 and the distance sensor 30 may be installed anywhere as long as the plant 500 can be captured within the viewing angle, but it is desirable to install the camera 20 and the distance sensor 30 at a position near the plant 500 in the vertical direction.

  As described above, the derivation of the water stress evaluation value and the derivation of the leaf increase / decrease evaluation value are not limited to the configuration in which the camera is used, and the derivation of the water stress evaluation value is evaluated using another device. May be. For example, by using the distance sensor 30 as another device, the leaf wrinkle caused by the water stress is measured as a height fluctuation, and the water stress state is evaluated by comparing the height fluctuation with a predetermined standard such as a threshold value. be able to.

  Next, a plant cultivation support method using the plant cultivation support system 2 will be described. FIG. 5 is an example of a flowchart showing processing of the plant cultivation support apparatus according to the fourth embodiment. The process of the plant cultivation support system 2 will be described in detail with reference to FIG.

  First, in step S <b> 201, the first evaluation unit 121 of the CPU 12 acquires distance data of the plant 500 from the distance sensor 30 via the interface 11 at a predetermined timing. The predetermined timing is, for example, the timing at which the distance sensor 30 receives a trigger signal from the CPU 12 via the interface 11. Alternatively, the timing may be determined based on a preset time or time interval.

  In step S201, distance data from a predetermined position (position where the distance sensor 30 is installed) to at least one point of the plant 500 may be obtained. That is, the distance data may be acquired from only one point in space or may be acquired from a plurality of points. Further, the distance data acquired from a plurality of points in the space may be integrated and converted into a representative value (average value, median value, standard deviation, etc.) for use. Hereinafter, a case where distance data is acquired from a plurality of points in space will be described as an example.

  Next, in step S202, the plant cultivation assistance apparatus 10 estimates a water stress state based on the distance data acquired in step S201. Specifically, the first evaluation unit 121 of the CPU 12 analyzes the distance data obtained from the distance sensor 30, estimates the water stress state of the plant 500 based on the distance data, and derives the evaluation value of the water stress. Examples of the method for estimating the moisture stress state from the distance data include the following.

  The first example is estimation based on a calibration curve. In other words, the moisture stress state is measured in advance by a method such as fracture measurement, and this is correlated with the distance data measured for the same individual at the same time, and a calibration curve is drawn. By using this calibration curve, the water stress state can be estimated from the distance data of an individual at an arbitrary time.

  The second example is estimation based on the amount of change in distance. That is, the representative value D0 of the distance data in a state where no wilting has occurred is extracted from the distance data in a predetermined time range before the time when the moisture stress state is estimated. By comparing the representative value D0 with the representative value Dt of the distance data at an arbitrary time, a deflation rate representing the degree of deflation of the leaves is calculated. Examples of the comparison method include a method for obtaining a ratio between D0 and Dt and a method for obtaining a difference between D0 and Dt. Based on this wrinkle rate, for example, a water stress state can be estimated using a calibration curve created in the same procedure as described above.

  The third example is estimation based on a three-dimensional shape. That is, based on the spatially acquired distance data, the shape of the whole plant and the shape of the leaf are three-dimensionally restored, and the water stress state is estimated using the three-dimensional shape. In this case, it is necessary to acquire distance data as densely as possible so that the shape of the whole plant and the shape of the leaves can be three-dimensionally restored.

  When using the shape of the whole plant, the reference shape of the whole plant in a state where no wilt has occurred is stored, and the degree of leaf wilt can be estimated based on the degree of change from the reference shape. As an example, based on the width and volume of the canopy in a state where no wilting has occurred as a reference, a method of converting the ratio of the width and volume of the crown at the current time into a degree of leaf wilting using a calibration curve prepared in advance. Is mentioned.

  When the leaf shape is used, the inclination of the leaf with respect to the horizontal plane can be obtained, and the degree of leaf wilting can be estimated based on the magnitude of the inclination.

  When using the shape of the whole plant or the shape of the leaf, the state of water stress is estimated from the obtained degree of leaf wilt and the correspondence between the preset leaf wilt and water stress. can do.

  Note that the water stress state estimated as described above may be an index serving as a measure of the water stress of the plant, and it is not always necessary to obtain the exact value of the water stress applied to the plant.

  In addition to plants, the distance data obtained from the distance sensor 30 may include not only the ground but also agricultural supplies such as support pillars and attraction strings as noise. If the moisture stress state is estimated using the distance data mixed with noise, a large error may occur in the estimation result. Therefore, it is desirable to remove these noises as appropriate.

  For example, since the distance to the ground can be calculated on the ground based on the height of the distance sensor, distance data corresponding to the ground can be removed. Agricultural supplies can be removed by taking the difference between the distance data acquired in the absence of a plant in advance and the distance data including the plant.

  Furthermore, an object that is not a plant may be specified from an image photographed by a camera based on color or shape characteristics, and data corresponding to the object in the distance data may be selectively excluded. This method can be adopted when a camera is used as the distance sensor 30, but image data from the camera 20 can be used when a camera is not used as the distance sensor 30.

  Next, in step S <b> 203, the second evaluation unit 122 of the CPU 12 acquires image data of the plant 500 from the camera 20 through the interface 11 at a predetermined timing. The predetermined timing is, for example, the timing at which the camera 20 receives a trigger signal from the CPU 12 via the interface 11. Alternatively, the timing may be determined based on a preset time or time interval.

  Next, in step S204, the plant cultivation support apparatus 10 derives an evaluation value of increase / decrease of the leaves of the plant 500 based on the image data acquired in step S203. Specifically, the second evaluation unit 122 of the CPU 12 analyzes the image data obtained from the camera 20 and evaluates the increase or decrease in the amount of leaves reflected in the image data. Then, based on the increase / decrease in the amount of leaves, it is estimated whether or not there has been a management work such as a leaf writing work or an attracting work. For the detection of the management work, the methods exemplified in the first to third embodiments can be used.

  Next, in step S <b> 205, the plant cultivation support device 10 determines whether or not liquid supply is necessary based on the evaluation value of moisture stress and the evaluation value of increase / decrease in leaves. Specifically, the control unit 123 of the CPU 12 determines whether or not the liquid supply to the plant 500 is necessary, the water stress evaluation value derived by the first evaluation unit 121 in step S202, and the second evaluation unit 122 in step S204. Judgment is made based on the evaluation value of the increase or decrease of the leaves derived from. The necessity of liquid supply is, for example, that the evaluation value of water stress exceeds a preset threshold value, and the evaluation value of leaf increase / decrease exceeds a preset threshold value (no management operation has been detected) ) Can be determined based on that.

  It should be noted that the water stress evaluation value may be updated based on the leaf increase / decrease evaluation value and then compared with the threshold value. Specifically, when it is determined that the management work has occurred based on the evaluation value of the increase or decrease of the leaf, the evaluation value of the water stress before the management work occurs or the water stress A method of overwriting with the evaluation value of the moisture stress in the absence state is conceivable.

  If it is determined in step S205 that liquid supply is necessary (Yes), the process proceeds to step S206. If it is determined that liquid supply is not required (No), the process proceeds to step S207.

  In step S206, the notification unit 70 performs a predetermined notification. That is, when the control unit 123 determines that liquid supply is necessary, the CPU 12 transmits information to the notification unit 70 and notifies the notification unit 70 that liquid supply is necessary. A specific example of the notification that liquid supply is necessary is as described above.

  The above steps S201 to S206 are a series of processes performed in association with the acquisition of one distance data and image data. When these steps are completed, it is determined in step S207 whether or not to end the control. If it is determined in step S207 that the process is to be ended (in the case of Yes), the control is ended, and if it is determined that the process is not to be ended (in the case of No), the process returns to step S201. The processing of S201 to S206 is repeated.

  In addition, although the example which estimates the presence or absence of management work based on the evaluation value of increase / decrease in the amount of leaves was shown in step S103, the reason for the increase / decrease in the amount of leaves is similar to the first embodiment. This is just an example. As described above, an increase or decrease in the amount of leaves leads to a misjudgment of the water stress state, so capturing the increase or decrease in the amount of leaves is the essence of the present invention, and estimating the occurrence of management work is the present invention. Is not the essence of

  In step S206, when it is determined that liquid supply is necessary, the contact information registered in advance is notified, but the present invention is not limited to this. For example, liquid supply means (not shown) may be used to automatically supply liquid to a plant by a preset amount according to the moisture stress state.

  According to the plant cultivation support apparatus according to the fourth embodiment, by combining the water stress evaluation value and the leaf increase / decrease evaluation value derived using various methods not limited to cameras, whether or not liquid supply is necessary False judgment can be suppressed. Other effects are the same as those in the first embodiment.

  The distance data can be used as image data. Specifically, if the distance data is two-dimensional data, the distance data can be handled as image data by regarding the distance to each point as luminance. In that case, the camera 20 becomes unnecessary, and step S203 in FIG. 5 can be omitted.

  The preferred embodiment has been described in detail above. However, the present invention is not limited to the above-described embodiment, and various modifications and replacements are made to the above-described embodiment without departing from the scope described in the claims. Can be added.

1, 2 Plant cultivation support system 10 Plant cultivation support device 11 Interface 12 CPU
DESCRIPTION OF SYMBOLS 13 Memory | storage part 14 Bus | bath 20 Camera 30 Distance sensor 70 Notification part 80 Storage medium mounting part 121 1st evaluation part 122 2nd evaluation part 123 Control part 500 Plant

Japanese Patent Laid-Open No. 2007-306846

Claims (12)

A plant cultivation support device for evaluating the state of a plant and supporting cultivation of the plant,
A first evaluation unit that obtains predetermined data related to the plant and derives an evaluation value of water stress of the plant based on the predetermined data;
A second evaluation unit that obtains image data of the plant and derives an evaluation value of increase or decrease in leaves of the plant based on the image data;
A control unit that determines a control to be applied to the plant based on the evaluation value of the water stress and the evaluation value of the increase or decrease of the leaves, and generates a control signal used for the control, Plant cultivation support device.   The second evaluation unit extracts a leaf region from the image data, derives a representative value of the leaf region, and evaluates an increase or decrease of the leaf based on a change amount of the representative value in a predetermined time interval. The plant cultivation support device according to claim 1, wherein the plant cultivation support device is derived.   The plant cultivation support device according to claim 2, wherein the representative value includes at least one of an area of the leaf region and a gravity center position of the leaf region.   The second evaluation unit extracts a leaf region from the image data, and evaluates the increase or decrease of the leaf based on a difference between the leaf region at a first time and the leaf region at a second time. The plant cultivation support apparatus according to claim 1, wherein a value is derived.   The plant cultivation support apparatus according to any one of claims 1 to 4, wherein the predetermined data is the image data obtained by the second evaluation unit.   The plant cultivation support device according to any one of claims 1 to 4, wherein the predetermined data is distance data from a predetermined position to at least one point of the plant. The plant cultivation support device according to any one of claims 1 to 6,
An image data providing means for providing the image data to the plant cultivation support device. A plant cultivation support device according to claim 6;
A plant cultivation support system, comprising: a distance sensor that acquires the distance data and provides the plant cultivation support device.   The plant cultivation support system according to claim 8, wherein the plant cultivation support apparatus handles the distance data as the image data. A plant cultivation support method for evaluating the state of a plant and supporting cultivation of the plant,
Obtaining predetermined data relating to the plant, and deriving an evaluation value of water stress of the plant based on the predetermined data;
Obtaining image data of the plant, and deriving an evaluation value of increase or decrease of leaves of the plant based on the image data;
And determining a control to be performed on the plant based on the evaluation value of the water stress and the evaluation value of the increase / decrease of the leaf, and generating a control signal used for the control. Plant cultivation support device.   The program which makes a computer perform each step of Claim 10.   A storage medium storing the program according to claim 11.

Images