JP7616656B2 - Strawberry plant leaf area calculation system - Google Patents

Description

The present invention relates to a system for calculating the leaf area of strawberry plants.

Total leaf area is an important index for diagnosing the growth of fruit vegetables and predicting yields, and can be used to estimate the amount of photosynthesis. In recent years, methods for measuring total leaf area from photographs of plants have been developed.

For example, in the case of tomatoes, cucumbers, peppers, etc., the cultivation method of hanging the stems from above is widespread, and each leaf can be observed from the outside, so measuring the total leaf area using images is relatively easy.

JP 2019-037225 A JP 2019-193581 A JP 2020-156431 A

However, in the case of strawberries, it is difficult to observe the leaves inside the plant from the outside, and the leaves of adjacent plants overlap, making them difficult to observe, so it is difficult to measure the total leaf area of each plant even using images taken from the outside.

The present invention aims to provide a strawberry plant leaf area calculation system that can accurately calculate the total leaf area of each strawberry plant.

The strawberry plant leaf area calculation system of the present invention comprises a blowing device that blows wind onto the strawberry plant from above, an image capture device that captures an image of the plant from above while the wind is blowing onto the plant, and a calculation device that calculates the total leaf area of the strawberry plant from the image. The calculation device has a memory unit that stores the leaf area of each leaf of the plant that was previously calculated, identifies target leaves that include at least the youngest first leaf and the next youngest second leaf of the plant from a newly captured first image, calculates the leaf area of the target leaves from the image, and calculates the total leaf area of the plant by adding up the leaf areas of the leaves other than the target leaves stored in the memory unit and the calculated leaf area of the target leaves.

The strawberry plant leaf area calculation system of the present invention has the effect of being able to accurately calculate the total leaf area of each strawberry plant.

FIG. 1 is a diagram illustrating a schematic configuration of a leaf area calculation system according to an embodiment. FIG. 2 is a diagram illustrating a hardware configuration of an information processing device. FIG. 2 is a functional block diagram of an information processing device. Figure 4(a) is an image of strawberry plants photographed from above, Figure 4(b) is an image of a single plant photographed from above, Figure 4(c) is an image photographed after wind was blown from above on the plant in Figure 4(b), and Figure 4(d) is a diagram to explain the effect of blowing wind. FIG. 13 is a diagram showing data stored in a leaf area DB. FIG. 13 is a diagram showing data stored in a total leaf area DB. 13 is a flowchart showing a first calculation process. 8(a) to 8(c) are diagrams for explaining a method for identifying the center of a stock. 9A is a diagram for explaining step S16 in FIG. 7, and FIG. 9B is a diagram for explaining step S18 in FIG. FIG. 13 is a diagram showing an example of a leaf area DB after a first calculation process has been performed. 13 is a flowchart showing the second and subsequent calculation processes. 12 is a flowchart showing detailed processing of step S52 in FIG. 11 . 1 is a graph showing the change in leaf area per week from the first leaf (new leaf) to the fifth leaf. 14(a) and 14(b) are diagrams showing changes in the leaf area DB when the process of step S60 in FIG. 11 is executed. 15(a) and 15(b) are diagrams showing changes in the leaf area DB when the process of step S62 in FIG. 11 is executed. 12 is a diagram showing a change in a leaf area DB when the process of step S64 in FIG. 11 is executed. FIG.

The leaf area calculation system according to one embodiment is described in detail below.

Figure 1 shows a schematic configuration of a leaf area calculation system 100 according to one embodiment. In this embodiment, the leaf area calculation system 100 is a system that calculates and records the leaf area (total leaf area) of each strawberry plant cultivated in an elevated cultivation bed 12 as shown in Figure 1. Note that the strawberry plants do not have to be cultivated in the cultivation bed 12, and may be planted in the ground in ridges, for example.

The leaf area calculation system 100 includes a monitoring device 10 and an information processing device 90 as a calculation device. The monitoring device 10 is a device that moves near the cultivation bed 12 and acquires images of strawberry plants taken from above. The information processing device 90 is communicably connected to the monitoring device 10, and calculates the leaf area (total leaf area) of each strawberry plant planted in the cultivation bed 12 based on images and position information acquired from the monitoring device 10. The information processing device 90 and the monitoring device 10 may be connected by wired connection such as a wired LAN (Local Area Network) or wirelessly such as WiFi. The information processing device 90 may be mounted on the monitoring device 10. In this embodiment, the extension direction of the cultivation bed 12 is the X-axis direction, the direction perpendicular to the X-axis direction in the horizontal plane is the Y-axis direction, and the vertical direction is the Z-axis direction.

As shown in FIG. 1, the monitoring device 10 is installed on rails 14 laid along the X-axis direction below the cultivation bed 12 (on the ground). The monitoring device 10 is movable in the X-axis direction along the rails 14.

The monitoring device 10 comprises a housing 20, wheels 22, a camera 24 as an imaging device, a blower 26 as a spraying device, a motor 28, a position detection device 29, and a control device 30. The housing 20 has a generally rectangular frame shape. The ceiling plate 20a of the housing 20 extends above the cultivation beds 12 and the strawberry stalks, and the underside (-Z surface) of the ceiling plate 20a faces the strawberry stalks vertically.

The wheels 22 are driven to rotate on the rails 14 by a motor 28. The rotation of the motor 28 is controlled by a control device 30.

The camera 24 and blower 26 are mounted on the underside of the ceiling panel 20a of the housing 20. The camera 24 photographs the strawberry stalks from above. The blower 26 has the function of blowing air from above to below the strawberry stalks. The control device 30 blows air onto the stalks using the blower 26, and in this state, uses the camera 24 to photograph the stalks from above. The image taken by the camera 24 is sent to the control device 30.

The position detection device 29 may be, for example, a position detection device having an RFID reader capable of communicating with multiple RFID (radio frequency identifier) tags installed near the cultivation bed 12, a position detection device having a camera capable of photographing a marker installed near the cultivation bed 12, or a position detection device that detects the position from the amount of rotation of the wheels 22 or the motor 28. The position detection device 29 detects the XY position of the camera 24. From the XY position of the camera 24, the XY position of each point in the captured image (capture range) can be specified. In addition, the detection result of the position detection device 29 is transmitted to the control device 30, and the control device 30 controls the motor 28 based on the detection result of the position detection device 29 to adjust the position of the monitoring device 10 (the capture location of the camera 24). In addition, the control device 30 transmits the image captured by the camera 24 and the position information (XY position) of the image to the information processing device 90.

The information processing device 90 uses the image and position information transmitted from the control device 30 to identify the position of the plant and calculate the total leaf area of each plant. FIG. 2 shows the hardware configuration of the information processing device 90. As shown in FIG. 2, the information processing device 90 includes a CPU (Central Processing Unit) 190, a ROM (Read Only Memory) 192, a RAM (Random Access Memory) 194, a storage unit (such as a HDD (Hard Disk Drive) or an SSD (Solid State Drive)) 196, a network interface 197, and a portable storage medium drive 199. These components of the information processing device 90 are connected to a bus 198. In the information processing device 90, the CPU 190 executes a program stored in the ROM 192 or the storage unit 196, or a program read by the portable storage medium drive 199 from the portable storage medium 191, thereby realizing the functions of the components shown in FIG. 3. The functions of each unit in FIG. 3 may be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

Figure 3 is a block diagram showing functions realized by the CPU 190 of the information processing device 90 executing a program. As shown in Figure 3, the information processing device 90 functions as an image acquisition unit 50, an image analysis unit 52, a data management unit 54, and a total leaf area calculation unit 56. Note that Figure 3 also shows a leaf area DB 60 and a total leaf area DB 62 stored in the memory unit 196.

The image acquisition unit 50 acquires from the control device 30 images of the plants photographed from above while wind is blowing on them by the blower 26. The image acquisition unit 50 also acquires location information of the images along with the images.

Here, the reason why the strawberry plant is photographed while wind is blown from the blower 26 in this embodiment will be described. FIG. 4(a) shows an image of a strawberry plant photographed from above. As shown in FIG. 4(a), it is difficult to observe the leaves on the inside of a strawberry plant from the outside, or the leaves of adjacent plants overlap, making it difficult to observe. FIG. 4(b) shows an image of one plant photographed from above, and FIG. 4(c) shows an image photographed after blowing wind from above the plant in FIG. 4(b). As can be seen by comparing FIG. 4(b) and FIG. 4(c), blowing wind from above onto the plant and photographing it makes it easier to photograph the leaves on the inside. For example, as shown in FIG. 4(d), it is easier to photograph the young leaves (the youngest first leaf (new leaf), the next youngest second leaf, the next youngest third leaf, etc.) that are present near the center of the plant.

Returning to FIG. 3, the image analysis unit 52 analyzes the image acquired by the image acquisition unit 50 to identify the center position of the plant present in the image, and identifies, for example, the first leaf (new leaf), second leaf, and third leaf as leaves to be measured. The image analysis unit 52 also analyzes the image to calculate the leaf area of the identified leaves to be measured (first leaf to third leaf).

The data management unit 54 obtains the analysis results (e.g., the calculation results of the leaf area of the measurement target leaf) by the image analysis unit 52, and updates the leaf area DB 60. Here, it is assumed that the leaf area DB 60 manages data such as that shown in FIG. 5 for each plant. As shown in FIG. 5, the leaf area DB 60 stores information such as "plant ID," "plant center position," "angle of the line connecting the first leaf and the plant center," and "leaf area of each leaf." The plant ID is identification information that is assigned to each plant immediately after the plant is planted. Details of the other information will be described later.

The total leaf area calculation unit 56 calculates the leaf area (total leaf area) of each plant using the information stored in the leaf area DB 60. The total leaf area calculation unit 56 stores the calculated total leaf area information in the total leaf area DB 62. In the total leaf area DB 62, for example, as shown in FIG. 6, the total leaf area value of each plant is stored in association with the plant ID for each calculation date.

(Regarding the processing of the information processing device 90)
Next, the processing of the information processing device 90 will be described in detail.

(Regarding the first calculation process)
First, the process for initially calculating the total leaf area of each plant after planting (first calculation process) will be described with reference to the flowchart in Fig. 7. The timing of this first calculation process is assumed to be the timing when four or more leaves have not yet appeared on the plant.

The monitoring device 10 photographs all the plants on the cultivation bed 12 in sequence. That is, the control device 30 photographs images using the camera 24 while moving the position of the monitoring device 10 (position of the camera 24) a specified distance in the X-axis direction based on the detection results of the position detection device 29. When photographing, the blower 26 blows air onto the strawberry plants from above. The images photographed by the camera 24 and position information of the images are sequentially transmitted from the control device 30 of the monitoring device 10 to the information processing device 90.

When the process in FIG. 7 is started, first, in step S10, the image acquisition unit 50 waits until an image and image location information are transmitted from the control device 30 of the monitoring device 10. When the image and location information are transmitted from the control device 30, the image acquisition unit 50 acquires the image and location information and proceeds to step S12.

When proceeding to step S12, the image analysis unit 52 detects the center of the plant present in the acquired image. For example, the image analysis unit 52 performs deep learning using a large number of images showing the shape of the plant and the center of the plant as learning data, and can detect the center of the plant present in the acquired image using the obtained learning model. The image analysis unit 52 can also detect the center of the plant using a method other than deep learning. For example, as shown in FIG. 8(a), the image analysis unit 52 can detect multiple petioles and detect the intersection of their extension lines as the center of the plant. By adopting this method, even if the center of the plant is hidden by the leaves of other plants as shown in FIG. 8(b), the center of the plant can be detected with high accuracy. Also, as shown in FIG. 8(c), strawberries usually have three leaves connected to one petiole, so the image analysis unit 52 can, for example, perform a process of drawing a straight line (see the dashed line in FIG. 8(c)) in the longitudinal direction of the middle leaf of the three leaves multiple times, and determine the intersection of these straight lines as the center of the plant. The image analysis unit 52 may preferentially detect the center of the plant using deep learning, and if the center of the plant cannot be detected by deep learning, may detect the center of the plant using the method of FIG. 8(a), FIG. 8(b), or FIG. 8(c). The image analysis unit 52 transmits the XY coordinates of the detected plant center to the data management unit 54, and the data management unit 54 stores the received XY coordinates in the leaf area DB 60. Specifically, the data management unit 54 assigns a plant ID to the plant whose center has been detected, stores the assigned plant ID in the "Plant ID" column of the leaf area DB 60 in FIG. 5, and stores the XY coordinates of the detected plant center in the "Plant center position" column.

Returning to FIG. 7, in the next step S14, the image analysis unit 52 executes a process of detecting the first leaf. For example, the image analysis unit 52 performs deep learning using a large number of learning data that associates the image of the leaf with the degree of unfolding to obtain a learning model. Then, the image analysis unit 52 can detect the unfolded leaves of the plant from within the image using the obtained learning model, and determine the detected unfolded leaves as the first leaf. The image analysis unit 52 may also detect leaves from the image and determine the smallest leaf among the detected leaves as the first leaf. The image analysis unit 52 may also detect leaves from the image and determine the leaf with the lightest color among the detected leaves as the first leaf. The image analysis unit 52 may also detect leaves from the image and determine the leaf with the shortest distance from the center of the plant as the first leaf. The image analysis unit 52 may preferentially apply a method of identifying the first leaf by deep learning, and if the first leaf cannot be identified by this method, the first leaf may be identified by the other method described above. The image analysis unit 52 may also identify the first leaf using multiple of the above-mentioned methods and combine the multiple identification results to identify the first leaf.

Next, in step S16, the image analysis unit 52 detects the angle of the line connecting the first leaf and the center of the plant with respect to the reference line. Specifically, as shown in FIG. 9(a), the image analysis unit 52 detects the angle (clockwise angle) θ between the line L connecting the center of the plant and the point P on the first leaf that is farthest from the center of the plant, and the reference line of the image (e.g., the Y axis). The image analysis unit 52 transmits the detected angle θ to the data management unit 54, and the data management unit 54 stores the received angle θ in the leaf area DB 60. Specifically, the data management unit 54 stores the angle θ in the "Angle of the line connecting the first leaf and the center of the plant" column in the leaf area DB 60 in FIG. 5.

9(a), the straight line L is a straight line connecting the point P of the first leaf farthest from the center of the plant and the center of the plant, but this is not limited to this. For example, the straight line L may be a straight line connecting the center of gravity of the image of the first leaf and the center of the plant. The straight line L may be the direction in which the petiole of the first leaf extends. A rectangle (approximate rectangle) surrounding the central leaf of the first leaf (usually the first leaf has three leaves, so the leaf located in the middle of them) may be specified, and the long side of the approximate rectangle may be the straight line L. A close rectangle (approximate rectangle) surrounding the entire first leaf (all three leaves) may be specified, and one of the sides of the approximate rectangle (for example, a side extending in a direction approximating the direction in which the petiole extends) may be the straight line L. Furthermore, the long axis of an ellipse approximating the central leaf of the first leaf, or the diagonal direction when the first leaf is approximated to a rhombus may be the straight line L. In other words, the straight line L may be any straight line other than those mentioned above, so long as it can be regarded as a straight line obtained from the image and can be regarded as the direction in which the first leaf grows from the center of the plant.

Next, in step S18, the image analysis unit 52 identifies the second and third leaves, and sets the first to third leaves as the leaves to be measured. Here, in the case of strawberries, as shown in FIG. 9(b), it is known that the second leaf is located at a point rotated about 144° from the direction in which the first leaf grows, starting from the center of the stalk. It is also known that the third leaf is located at a point rotated about 144° from the direction in which the second leaf grows, starting from the center of the stalk (at a point rotated about 288° from the direction in which the first leaf grows). Therefore, the image analysis unit 52 identifies the second and third leaves based on the first leaf, and sets the first to third leaves as the leaves to be measured. Note that there may be cases in which the third leaf does not exist, in which case the first and second leaves may be the leaves to be measured.

Next, in step S20, the image analysis unit 52 calculates the leaf area of the measurement target leaf. In this case, the image analysis unit 52 can calculate the leaf area of the measurement target leaf using the number of pixels (area) corresponding to the measurement target leaf and a predetermined shooting height. The shooting height may be calculated using images from multiple cameras, or may be measured using a depth camera or distance sensor. In addition, the height to the plant may be three-dimensionalized from multi-point measurements by mobile measurement. The image analysis unit 52 transmits the measured leaf area to the data management unit 54, and the data management unit 54 stores the received leaf area in the leaf area DB 60. Specifically, the data management unit 54 stores the measured leaf area in the leaf area column of each leaf in the leaf area DB 60 in FIG. 5.

Next, in step S22, the total leaf area calculation unit 56 adds up the leaf areas of all the leaves stored in the leaf area DB 60 and stores the total leaf area in the total leaf area DB 62. This completes the first calculation process for one plant. If another plant is present in the acquired image, the processes of steps S12 to S22 in FIG. 7 are repeated for that plant. Even after the process of FIG. 7 is completed, the process of FIG. 7 is repeated for images captured sequentially by the monitoring device 10, and the total leaf area of all the plants being cultivated in the cultivation bed 12 is calculated.

The process in FIG. 7 (first calculation process) may be performed when four or more leaves appear on the plant.
For example, (1) the leaf area of the first leaf (new leaf), second leaf, and third leaf may be calculated using the process of Figure 7, and the leaf area of the other leaves may be calculated from images taken from above, and these may be added together to calculate the total leaf area of the plant.
Also, for example, (2) the leaf areas of the first leaf (new leaf), second leaf, and third leaf may be calculated using the process of Figure 7, and the leaf areas of the other leaves may be assumed to be predetermined values, and these may be added together to calculate the total leaf area of the plant.
Furthermore, for example, (3) the leaf areas of the first leaf (new leaf), second leaf, and third leaf may be calculated by the process in Fig. 7, and the leaf areas of the other leaves may be ignored (leaf area = 0). This is because if the first calculation process is close to the time of planting, the size of the other leaves is not so large that it is considered that there is no problem in ignoring them.

(Regarding calculation process from the second time onwards)
Next, the second and subsequent leaf area calculation processes are performed according to the flowcharts of Figures 11 and 12. In the following description, the second calculation process of a stock (stock ID = 00001) in which data such as that shown in Figure 10 was stored in the leaf area DB 60 in the first calculation process will be described. Note that the timing at which the second calculation process is performed is, for example, a timing at which the leaf identified as the first leaf (new leaf) in the first calculation process is identified as the second leaf or the third leaf.

In the process of FIG. 11, first, in step S50, the image acquisition unit 50 waits until an image and image location information are transmitted from the control device 30 of the monitoring device 10. When the image and location information are transmitted from the control device 30, the image acquisition unit 50 acquires them and proceeds to step S52.

When the process proceeds to step S52, the image analysis unit 52 executes a process for detecting the first leaf. Specifically, the image analysis unit 52 executes a process in accordance with the flowchart in FIG. 12.

In the process of FIG. 12, the image analysis unit 52 first detects the center of the stock in step S102. This process is the same as step S12 described above. The image analysis unit 52 then refers to the leaf area DB 60 to identify the data (stock ID) that stores the center coordinates closest to the center coordinates of the detected stock. Here, since the center of the stock with stock ID=00001 shown in FIG. 10 is closest to the center of the stock detected this time, stock ID=00001 is identified.

Next, in step S104, the image analysis unit 52 detects the first leaf. This process is similar to step S14 described above.

Next, in step S106, the image analysis unit 52 detects an angle (assumed to be angle _θ2 ) between the line connecting the first leaf and the center of the plant and the reference line. This process is similar to that of step S16 described above.

Next, in step S108, the image analysis unit 52 calculates the difference between the angle ( _θ1 in FIG. 10) of the straight line connecting the first leaf and the center of the plant, obtained in the previous measurement, and the reference line. In this case, the image analysis unit 52 calculates _θ1 - _θ2 .

In step S110, the image analysis unit 52 judges whether the difference (θ ₁ - θ ₂ ) calculated in step S108 is approximately 0°. If the judgment in step S110 is positive, the process proceeds to step S112, where the image analysis unit 52 specifies that the detected first leaf is the same as the first leaf detected previously. After that, all the processes in FIG. 12 are terminated, and the process proceeds to step S54 in FIG. 11.

On the other hand, if the determination in step S110 is negative, the process proceeds to step S114, where the image analysis unit 52 determines whether the difference is approximately 144°. If the determination in step S114 is positive, the process proceeds to step S116, where the image analysis unit 52 identifies the detected first leaf as the first newly emerged leaf. After that, all the processing in FIG. 12 is terminated, and the process proceeds to step S54 in FIG. 11.

On the other hand, if the determination in step S114 is negative, the process proceeds to step S118, where the image analysis unit 52 determines whether the difference is approximately 288°. If the determination in this step S118 is positive, the process proceeds to step S120, where the image analysis unit 52 identifies the detected first leaf as the newly emerged second leaf. After that, all the processes in FIG. 12 are terminated, and the process proceeds to step S54 in FIG. 11.

On the other hand, if the determination in step S118 is negative, the process proceeds to step S122, where the image analysis unit 52 outputs an error and ends the process without executing the subsequent processes in FIG. 11. Note that, since a negative determination in step S118 is considered to be a case where the measurement interval is too long or where side shoots have appeared, the image analysis unit 52 notifies the producer, etc., to that effect.

When the process of FIG. 12 is performed and the process proceeds to step S54 of FIG. 11, the image analysis unit 52 identifies the second and third leaves and sets the first to third leaves as the measurement target leaves. The process of step S54 is the same as step S18 of FIG. 7 described above, so the second and third leaves may be identified based on the first leaf. However, in the second and subsequent calculation processes, the information obtained in the previous calculation process may be used to check whether the second and third leaves have been correctly identified, thereby improving the accuracy of identification. For example, since the areas of the current second leaf (e.g., the previous first leaf) and the current third leaf (e.g., the previous second leaf) were obtained in the previous calculation process, the areas of the current second and third leaves may be estimated from the areas and the calculation process execution interval, and if the areas of the identified second and third leaves are similar to the estimated areas, it may be determined that the second and third leaves have been correctly identified. Also, for example, if the distance from the center of the plant to the current second leaf (e.g., the previous first leaf) and the current third leaf (e.g., the previous second leaf) was obtained in the previous calculation process, the distance from the center of the plant to the current second and third leaves can be estimated from that distance and the interval between calculation processes, and if the identified distance from the center of the plant to the second and third leaves is similar to the estimated distance, it can be determined that the second and third leaves have been correctly identified.

In addition, it is generally known that the order of the first leaf, second leaf, and third leaf is the largest in terms of (1) increase in leaf area, (2) degree of elongation of petiole length, and (3) change in color. Therefore, this information can be used to confirm whether the second and third leaves have been correctly identified.

Next, in step S56, the image analysis unit 52 calculates the leaf area of the measurement target leaf. This process is the same as step S20 in FIG. 7. Although the plant has leaves other than the measurement target leaves, the leaf area of the leaves other than the measurement target leaves is not calculated. Here, FIG. 13 shows the change in leaf area of the first leaf (new leaf) to the fifth leaf per week. As shown in FIG. 13, the leaf area of the first leaf (new leaf), second leaf, and third leaf changes significantly, while the leaf area of the fourth leaf and the fifth leaf hardly changes (grows). Therefore, in this embodiment, in the second and subsequent calculation processes, the leaf area of all the leaves of the plant is not calculated, but only the leaf area of the measurement target leaves (here, the first leaf to the third leaf) is calculated, and the leaf area obtained in the previous calculation process is used as it is for the fourth leaf and subsequent leaves.

Next, in step S58, the data management unit 54 judges whether the detected first leaf is a leaf that did not exist in the previous time. In this case, in step S52b, the data management unit 54 judges whether either step S116 or S120 in FIG. 12 has been passed. If the judgment in step S58 is positive, the process proceeds to step S60, where the image analysis unit 52 rewrites the leaf area DB 60 so that the number of newly detected leaves is m and the previous nth leaf becomes the n+mth leaf. For example, when one new leaf is detected (when step S116 in FIG. 12 has been passed), the leaf area DB 60 in FIG. 10 is rewritten as shown in FIG. 14(a). Also, when two new leaves are detected (when step S120 in FIG. 12 has been passed), the leaf area DB 60 in FIG. 10 is rewritten as shown in FIG. 14(b).

Next, in step S62, the data management unit 54 overwrites the leaf area of the measurement target leaf in the leaf area DB 60. For example, if the leaf area of the first leaf = S1a', the leaf area of the second leaf = S2a', and the leaf area of the third leaf = S3a' are calculated in step S56, then if the leaf area DB 60 is that of FIG. 14(a), it will be overwritten as shown in FIG. 15(a). Also, if the leaf area DB 60 is that of FIG. 14(b), it will be overwritten as shown in FIG. 15(b). Then, the process proceeds to step S66.

On the other hand, if the determination in step S58 is negative, i.e., if step S112 in FIG. 12 has been executed, the process proceeds to step S64, where the data management unit 54 overwrites the leaf area of the measurement target leaf in the leaf area DB 60. For example, the leaf area DB 60 in FIG. 10 is overwritten as shown in FIG. 16. Then, the process proceeds to step S66.

Next, when the process proceeds to step S66, the data management unit 54 identifies leaves to be subtracted from the leaf area DB 60 and deletes them from the leaf area DB 60. For example, if the producer has recorded which leaves were removed or which leaves died, the data management unit 54 deletes the data of the removed leaves and the dead leaves from the leaf area DB 60. Also, if the number of leaves to be left for each plant is predetermined depending on the season, the data management unit 54 deletes data of old leaves from the leaf area DB 60 based on the period from the start of cultivation. Also, since old leaves have low photosynthetic activity, the data management unit 54 deletes data of old leaves from the leaf area DB 60 so as not to include old leaves in the total leaf area. Furthermore, since the lower leaves overlap with other leaves and are less exposed to light, the data management unit 54 deletes data of the lower leaves from the leaf area DB 60.

Next, in step S68, the total leaf area calculation unit 56 calculates the leaf area of all leaves stored in the leaf area DB 60 (total leaf area) and stores it in the total leaf area DB 62. This process is the same as step S22 in FIG. 7.

This completes the second and subsequent calculation processes for one plant. If another plant is present in the acquired image, the processes in Figures 11 and 12 are repeated for that plant. Even after the process in Figure 11 is completed, the process in Figure 11 is repeated for images captured sequentially by the monitoring device 10, until the total leaf area of all plants being cultivated in the cultivation bed 12 is calculated.

The total leaf area of each plant calculated as described above can be used to estimate the amount of photosynthesis in each plant. Therefore, by using the total leaf area calculated with high accuracy as in this embodiment, the amount of photosynthesis can be estimated with high accuracy. When estimating the amount of photosynthesis, the leaf area of each leaf can be used instead of the total leaf area. In this case, by taking into account the difference in photosynthetic activity according to the leaf position in estimating the amount of photosynthesis, the amount of photosynthesis can be estimated with high accuracy. Furthermore, by using the amount of photosynthesis estimated with high accuracy, it is possible to perform growth prediction and yield prediction with high accuracy.

As described above in detail, the leaf area calculation system 100 of this embodiment includes a monitoring device 10 having a blower 26 that blows wind from above onto the strawberry plant, a camera 24 that takes an image of the plant from above while the wind is blowing onto the plant, and an information processing device 90 that calculates the total leaf area of the strawberry plant from the image. This information processing device 90 has a leaf area DB 60 that stores the leaf area of each leaf calculated in the past. The information processing device 90 then identifies the measurement target leaf (first to third leaves) from the newly captured image, calculates the leaf area of the measurement target leaf from the image, and calculates the total leaf area of the plant by adding up the leaf area of the leaves other than the measurement target leaf stored in the leaf area DB 60 and the calculated leaf area of the target leaf. In this way, in this embodiment, the plant is photographed from above while the blower 26 is blowing wind onto it, making it easier to photograph young leaves (e.g., first to third leaves) that exist near the center of the plant. In addition, when calculating the total leaf area, young leaves are used as the measurement target leaves, and only the leaf area of the measurement target leaves is calculated. For the leaf area of leaves other than the measurement target leaves (leaves whose leaf area hardly changes), the leaf area stored in the leaf area DB 60 is used, so the total leaf area of the strawberry plant can be calculated with high accuracy. In addition, because the leaf area of some leaves of the plant is not calculated, the amount of processing can be reduced compared to when the leaf area of all leaves is calculated.

In addition, in this embodiment, when calculating the total leaf area of a plant, the information processing device 90 does not add the leaf area of leaves that have been removed from the plant (defoliated leaves), leaves that are estimated to have been removed from the plant, or leaves that are estimated not to contribute to photosynthesis. This makes it possible to accurately calculate the total leaf area of a plant and the total leaf area of a plant that contributes to photosynthesis.

In addition, in this embodiment, the information processing device 90 analyzes the captured image to identify the center of the plant and the position of the first leaf (new leaf), and identifies the second and third leaves based on the straight line connecting the center of the plant and the first leaf. This makes it possible to accurately identify the second and third leaves by utilizing the characteristics of the strawberry plant.

In addition, in this embodiment, the information processing device 90 detects (Figures 8(a) and 8(b)) or estimates (Figure 8(c)) the positions of two or more petioles of the plant from the captured image, and identifies the position where the two or more petioles intersect as the center of the plant. This makes it possible to accurately identify the center of the plant by utilizing the characteristics of the strawberry plant. Furthermore, by using the identified plant center to manage the plant data by treating plants with the same center position as the same plant, the plant data can be managed continuously.

In the above embodiment, the measurement leaves are the first to third leaves, but this is not limited to the above. Although this will vary depending on the variety, cultivation method, and cultivation region, if it is determined in the graph of FIG. 13 that the amount of change in the third leaf is almost the same as that in the fourth and fifth leaves, the measurement leaves may be only the first and second leaves. Also, if it is determined in the graph of FIG. 13 that the amount of change in the fourth leaf is large, the measurement leaves may be the first to fourth leaves.

In the above embodiment, the monitoring device 10 moves along the rail 14, but the present invention is not limited to this. For example, an unmanned aerial vehicle (drone or multicopter) equipped with a camera 24, blower 26, position detection device 29, and control device 30 may be used as the monitoring device. In this case, if the wind pressure from the propellers of the unmanned aerial vehicle is equivalent to the wind pressure from the blower 26, the blower 26 may be omitted.

The above-described embodiment is a preferred example of the present invention. However, the present invention is not limited to this embodiment, and various modifications are possible without departing from the spirit of the present invention.

10 Monitoring device 24 Camera (photography device)
26 Blower (spraying device)
60 Leaf area DB (memory part)
90 Information processing device (calculation device)
100 Leaf area calculation system

Claims (7)

A blowing device that blows air onto strawberry plants from above,
A photographing device that photographs an image of the plant from above while wind is blowing on the plant;
A calculation device for calculating a total leaf area of the strawberry plant from the image,
The calculation device is
A memory unit that stores previously calculated leaf areas of each of the leaves of the plant,
Identifying target leaves including at least the youngest first leaf and the next youngest second leaf of the plant from the newly captured first image;
Calculating the leaf area of the target leaf from the image;
The leaf area of the leaves other than the target leaf stored in the storage unit and the calculated leaf area of the target leaf are summed to calculate a total leaf area of the plant.
A system for calculating the leaf area of strawberry plants. The strawberry plant leaf area calculation system according to claim 1, characterized in that the target leaves are the first leaf, the second leaf, and the third leaf, which is the next youngest leaf after the second leaf. The strawberry plant leaf area calculation system according to claim 1 or 2, characterized in that the calculation device does not add the leaf area of leaves that have disappeared from the plant, leaves that are estimated to have disappeared from the plant, and leaves that are estimated not to contribute to photosynthesis when calculating the total leaf area of the plant. The computing device analyzes the first image to identify a center of the plant and a location of the first leaf;
The strawberry plant leaf area calculation system according to any one of claims 1 to 3, characterized in that the target leaves other than the first leaf are identified based on a straight line connecting the center of the plant and the first leaf. The strawberry plant leaf area calculation system according to claim 4, characterized in that the calculation device detects or estimates the positions of two or more petioles of the plant from the first image, and identifies the position where the two or more petioles intersect as the center of the plant. The strawberry plant leaf area calculation system according to claim 4 or 5, characterized in that the calculation device analyzes the first image to detect unexpanded leaves and identifies the unexpanded leaves as the first leaves. The strawberry plant leaf area calculation system according to any one of claims 4 to 6, wherein the calculation device identifies the first leaf based on its positional relationship with the center of the plant or on characteristics of the leaf.

Images