Tea tree information estimation method and tea tree information estimation program

Abstract

To accurately estimate the sprout growth amount of a tea plant.SOLUTION: A three-dimensional information acquisition part of a server acquires three-dimensional information (DSM: fig. 5 (b)) of a tea plant in which a state after training or mechanical picking is maintained, and three-dimensional information (DSM: fig. 5 (a)) of the tea plant just before sprout harvesting. A growth amount estimation part of the server estimates the sprout growth amount of the tea plant based on the difference between the two pieces of three-dimensional information (fig. 5 (c)). This allows accurate estimation of the sprout growth amount just before harvest, based on the state of the plant after training or mechanical picking.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to a method for estimating tea tree information and a program for estimating tea tree information.

Background Art

[0002] 0>In the production and manufacturing of teas such as green tea and black tea, tea leaves (new shoots of tea trees) picked from tea trees in tea plantations are used. However, the new shoots of tea trees, which are the picking parts, grow daily, and the yield varies depending on the growth amount.

[0003] Conventionally, a technique has been known in which a vegetation index (NDVI: Normalized Difference Vegetation Index) is calculated using optical data included in image information of tea leaves, and the picking suitability of tea leaves is evaluated or the yield is estimated using the calculated vegetation index (see, for example, Patent Document 1, Non-Patent Documents 1, 2, etc.). Also, a technique is known in which the total cross-sectional area of cut branches after autumn pruning is derived by image analysis, and the yield is predicted based on the total cross-sectional area of the cut branches (see, for example, Non-Patent Document 3, etc.).

[0004] Also, techniques for estimating the yield of crops such as root vegetables from three-dimensional information of the field and techniques for predicting the yield using images of fruit vegetables, fruiting vegetables, leafy vegetables, etc. are known (see, for example, Patent Documents 2, 3, etc.).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

Non-Patent Documents

[0006]

Non-Patent Document 1

[0007] However, estimation processing using vegetation indices requires expensive equipment such as multispectral cameras. Furthermore, since the cultivation methods for tea differ from those for fruit vegetables, leafy vegetables, etc., it is not possible to accurately estimate the growth of new shoots even when using the techniques disclosed in Patent Documents 2 and 3.

[0008] The present invention aims to provide a method and program for estimating tea tree information that can accurately estimate the amount of new shoot growth in tea trees. [Means for solving the problem]

[0009] The present invention provides a method for estimating tea tree information, which involves obtaining first three-dimensional information showing the distribution of height information of the tea tree during a first period in which the state after pruning or after mechanical harvesting of tea leaves is maintained, and second three-dimensional information showing the distribution of height information of the tea tree during a second period following the first period, and then determining the tea tree based on the difference between the second three-dimensional information and the first three-dimensional information. And newly grown after the first period This is a method for estimating tea plant information, in which a computer performs the processing to estimate the amount of new shoot growth.

[0010] The tea tree information estimation program of the present invention acquires first three-dimensional information showing the distribution of height information of the tea tree during a first period in which the state after pruning or after mechanical harvesting of tea leaves is maintained, and second three-dimensional information showing the distribution of height information of the tea tree during a second period following the first period, and based on the difference between the second three-dimensional information and the first three-dimensional information, it estimates the tea tree And newly grown after the first period This program uses a computer to perform a process that estimates the growth rate of new shoots. [Effects of the Invention]

[0011] The tea tree information estimation method and tea tree information estimation program of the present invention have the effect of being able to accurately estimate the amount of new shoot growth in tea trees. [Brief explanation of the drawing]

[0012] [Figure 1] This figure shows the configuration of a tea tree information estimation system according to one embodiment. [Figure 2] Figure 2(a) shows an example of the hardware configuration of a user terminal, and Figure 2(b) shows an example of the hardware configuration of a server. [Figure 3] This is a functional block diagram of the user terminal and server. [Figure 4] This flowchart shows an example of server processing. [Figure 5] Figure 5(a) shows the Digital Surface Model (DSM) immediately before the first tea harvest, Figure 5(b) shows the DSM after pruning in the autumn and winter, and Figure 5(c) shows the difference between the DSMs in Figures 5(a) and 5(b). [Figure 6] This graph shows the relationship between harvesting length and yield. [Modes for carrying out the invention]

[0013] Hereinafter, an embodiment will be described in detail based on FIGS. 1 to 6. FIG. 1 schematically shows the configuration of a tea tree information estimation system 100 according to an embodiment. The tea tree information estimation system 100 of this embodiment is a system that can be used by producers who cultivate tea trees in a tea garden, pick tea leaves from the tea trees, and ship them.

[0014] As shown in FIG. 1, the tea tree information estimation system 100 includes a user terminal 70 that can be used by a producer, a drone 60 having a camera 62 and a GNSS (Global Navigation Satellite System) sensor 64 that can communicate with the user terminal 70, and a server 10. The user terminal 70 and the server 10 are connected to a network 80 such as the Internet, and information can be exchanged between the devices. Here, it is assumed that the producer flies the drone 60 over the tea garden and uses the camera 62 to photograph the tea garden from above. For example, when flying the drone 60, the altitude above the ground can be 20 m and the speed can be 4.0 m / s. Note that the drone 60 may fly over the tea garden in response to an operation of a controller by the producer, or may automatically fly over the tea garden along a preset flight route. Also, the photographing by the camera 62 may be performed in response to an operation of a controller by the producer, or may be automatically performed at a preset position and timing.

[0015] (User Terminal 70) The user terminal 70 acquires the image captured by the camera 62 and the position information of the camera 62 at the time of image capture measured by the GNSS sensor 64, and generates three-dimensional information of the tea plantation (see FIGS. 5(a) and 5(b)) from the acquired image and position information. The three-dimensional information is, for example, a DSM (Digital Surface Model). The DSM is a numerical elevation modeling of the surface layer that combines the ground (topography) and features such as tea trees on it. The DSM is generated using a plurality of photos (images) from above. However, it is not limited to this. When the drone 60 has a laser distance measuring device, the DSM may be generated using laser surveying technology. When generating the DSM using a plurality of images from above, in order to obtain a high-precision DSM, for example, the camera 62 mounted on the drone 60 is used to capture a plurality of images such that the overlap of the images (the overlapping rate of the images in the traveling direction of the drone) is 80% and the side overlap (the overlapping rate of the images in the direction perpendicular to the traveling direction of the drone (width direction)) is 60%.

[0016] FIG. 2(a) shows the hardware configuration of the user terminal 70. As shown in FIG. 2(a), the user terminal 70 includes a CPU (Central Processing Unit) 190, a ROM (Read Only Memory) 192, a RAM (Random Access Memory) 194, a storage (HDD (Hard Disk Drive) or SSD (Solid State Drive)) 196, a communication unit 197, a display unit 193, and an input unit 195. The user terminal 70 also includes a portable storage medium drive 199 that can read data and programs stored in the portable storage medium 191. Each component of the user terminal 70 is connected to a bus 198. The display unit 193 is a liquid crystal display, an organic EL display, etc., and the input unit 195 is a touch panel, etc. The user terminal 70 functions as an image / position information acquisition unit 20, a three-dimensional information generation unit 22, and a three-dimensional information transmission unit 24 shown in FIG. 3 when the CPU 190 executes a program.<>

[0017] The image and location information acquisition unit 20 acquires images captured by the camera 62 and location information of the camera 62 at the time of image capture, which is measured by the GNSS sensor 64. The image and location information acquisition unit 20 acquires images and location information from the camera 62 and GNSS sensor 64 via the communication unit 197 (see Figure 2(a)). However, the image and location information acquisition unit 20 is not limited to this, and may also acquire images captured by the camera 62 and location information measured by the GNSS sensor 64 using a portable storage medium 191.

[0018] The three-dimensional information generation unit 22 generates three-dimensional information (DSM) of the tea plantation using images and location information acquired by the image and location information acquisition unit 20. The producer will take photographs of the tea plantation at two different times, and the three-dimensional information generation unit 22 will generate three-dimensional information of the tea plantation for each of the two times. Here, the two times are, for example, the first period (after pruning in autumn and winter and before the sprouting of new shoots) and the second period (after the sprouting of new shoots and immediately before the first tea harvest). Pruning in autumn and winter is done to make the surface for harvesting the first tea of ​​the following year uniform and to prevent old leaves and stems from being mixed in with the fresh leaves of the next tea season. Therefore, the three-dimensional information generation unit 22 generates three-dimensional information from images taken at the first period that shows the distribution of height information (distribution in a two-dimensional plane) of the tea trees that have been pruned into an arc shape so that the surface is uniform. Furthermore, the three-dimensional information generation unit 22 generates three-dimensional information of the tea plant in a state where new shoots have grown, from images taken during the second period.

[0019] The three-dimensional information transmission unit 24 transmits the three-dimensional information generated by the three-dimensional information generation unit 22 to the server 10.

[0020] (Server 10) Returning to Figure 1, Server 10 acquires three-dimensional information from the user terminal 70 for two different time periods and estimates the amount of new shoot growth in the tea plantation from the acquired three-dimensional information. Server 10 also accepts input information on the height of tea leaf picking (information on how high to cut from the surface (plant face) after pruning) and estimates the tea leaf yield in the tea plantation based on the accepted picking height information and the information on the amount of new shoot growth in the tea plantation.

[0021] Figure 2(b) shows an example of the hardware configuration of server 10. As shown in Figure 2(b), server 10 includes a CPU 90, ROM 92, RAM 94, storage 96, communication unit 97, and a portable storage medium drive 99, etc. Each of these components of server 10 is connected to a bus 98. In server 10, the CPU 90 executes programs (including a tea tree information estimation program) stored in ROM 92 or storage 96, or programs read from the portable storage medium 91 by the portable storage medium drive 99, thereby realizing the functions of each component shown in Figure 3. Note that the functions of each component in Figure 3 may be realized by integrated circuits such as ASICs (Application Specific Integrated Circuits) or FPGAs (Field Programmable Gate Arrays).

[0022] Figure 3 shows a functional block diagram of server 10. In server 10, the CPU 90 executes a program, and as shown in Figure 3, it functions as a three-dimensional information acquisition unit 30, a growth rate estimation unit 32, a pruning height information acquisition unit 34, and a yield estimation unit 36.

[0023] The three-dimensional information acquisition unit 30 acquires three-dimensional information from two different time periods transmitted from the three-dimensional information transmission unit 24 of the user terminal 70.

[0024] The growth rate estimation unit 32 estimates how much the new shoots of tea plants in the tea plantation grew between the first and second periods by taking the difference between the three-dimensional information acquired by the three-dimensional information acquisition unit 30 for two periods.

[0025] The harvesting height information acquisition unit 34 acquires information (harvesting height information) on the height at which tea leaves are harvested from the plant surface when pruning is performed in the autumn and winter. The harvesting height information may be entered by the producer from the user terminal 70, or it may be set in advance.

[0026] The yield estimation unit 36 ​​estimates the yield of tea leaves in the tea plantation based on the growth rate estimated by the growth rate estimation unit 32 and the harvesting height information.

[0027] (Regarding the processing on Server 10) Next, the processing of server 10 will be explained in detail, following the flowchart in Figure 4, with reference to other diagrams as appropriate.

[0028] The process shown in Figure 4 begins when three-dimensional information for the same tea plantation at two different time periods (the first and second periods) is transmitted from the user terminal 70. When the process shown in Figure 4 begins, first, in step S10, the three-dimensional information acquisition unit 30 acquires three-dimensional information for the two time periods. For example, suppose that the three-dimensional information for the two time periods obtained is a DSM for the period immediately before the first tea harvest (the second period), as shown in Figure 5(a), and a DSM for the period after pruning in autumn and winter, before the sprouting of new shoots (the first period), as shown in Figure 5(b).

[0029] Next, in step S12, the growth rate estimation unit 32 calculates the difference between the three-dimensional information of the two time periods and estimates the growth rate. The difference between the three-dimensional information of the two time periods is as shown in Figure 5(c), which represents how much the new shoots have grown (growth rate) at each location in the tea plantation since the autumn / winter period.

[0030] Next, in step S14, the pruning height information acquisition unit 34 acquires pruning height information. For example, the pruning height information acquisition unit 34 acquires information that pruning is performed at a height S (m) from the surface of the plant after pruning in the autumn and winter.

[0031] Next, in step S16, the yield estimation unit 36 ​​calculates the difference between the harvesting height information and the growth amount and estimates the yield. For example, the yield estimation unit 36 ​​calculates the yield from Figure 5(c) for a certain range of the tea plantation (for example, 1m). 2 The range of the tea plantation is identified, and the average growth rate D(m) of the identified range is obtained. The yield estimation unit 36 ​​then subtracts the picking height information S(m) from the average growth rate D(m) of the identified range to obtain the picking length L(m) (=DS). The yield estimation unit 36 ​​then uses a graph like the one shown in Figure 6 (a graph showing the relationship between picking length L and yield H), which has been prepared in advance, to determine the yield H of the identified range. Each point in the graph in Figure 6 is a plot of the relationship between the measured value of picking length L obtained in advance and the measured value of yield at that time, and the dashed line is a straight line approximating each plot using the least squares method or the like. The graph in Figure 6 may be prepared for each tea plantation, or the same graph may be prepared for each region, each variety, etc., within a defined range where it can be used.

[0032] Subsequently, the yield estimation unit 36 ​​repeatedly performs the above process while varying the range to be identified, estimating the yield for each range. Then, the yield estimation unit 36 ​​estimates the total yield for the entire tea plantation by summing the yields for each range.

[0033] Furthermore, the three-dimensional information acquisition unit 30 may acquire orthomosaic images of the tea plantation along with the DSM. In this case, when the yield estimation unit 36 ​​sequentially identifies a certain range of the tea plantation in step S16, it can identify only the range corresponding to the green area of ​​the orthomosaic image. This allows the unit to identify only the range where tea trees exist (harvesting surface) when sequentially identifying a certain range, thereby enabling accurate estimation of the total yield of the tea plantation. Furthermore, when the three-dimensional information generation unit 22 generates three-dimensional information (DSM) of the tea plantation, it may include color information acquired by an RGB camera (visible light camera) in the three-dimensional information. By doing so, when the yield estimation unit 36 ​​sequentially identifies a certain range of the tea plantation, it can identify only the range corresponding to the green area based on the three-dimensional information.

[0034] Next, in step S18, the yield estimation unit 36 ​​outputs the growth rate estimated in step S12 and the yield estimated in step S16 to the user terminal 70. As a result, the user terminal 70 displays the estimated growth rate of new shoots and the yield in the tea plantation. By looking at the information displayed on the user terminal 70, producers can understand whether the tea plants in the tea plantation are growing well and how much tea leaves can be harvested from the plantation.

[0035] As described in detail above, according to this embodiment, the three-dimensional information acquisition unit 30 acquires three-dimensional information (DSM) of the tea plants during the period when the state after pruning in autumn and winter is maintained (first period), and three-dimensional information (DSM) of the tea plants during the period immediately before the first tea harvest (second period). The growth amount estimation unit 32 estimates the growth amount of new shoots in the tea plants based on the difference between the two sets of three-dimensional information. As a result, the growth amount estimation unit 32 can accurately estimate the growth amount of new shoots immediately before harvest, using the state after pruning in autumn and winter as a reference. Furthermore, in this embodiment, by using the DSM of the tea plantation from the first and second periods, the growth amount of new shoots can be estimated accurately regardless of the location or land shape of the tea plantation. In addition, since the vegetation index (NDVI) is not used to estimate the growth amount in this embodiment, there is no need to prepare expensive equipment such as a multispectral camera. As a result, the growth amount can be estimated simply and inexpensively.

[0036] Furthermore, according to this embodiment, the harvesting height information acquisition unit 34 acquires the harvesting height information S(m) of the tea leaves, and the yield estimation unit 36 ​​estimates the tea leaf yield based on the growth amount D(m) of the new shoots and the harvesting height information S(m). This makes it possible to accurately estimate the yield while taking into account the height (depth) at which the producer harvests the first flush tea leaves.

[0037] In the above embodiment, the three-dimensional information acquisition unit 30 was described as acquiring DSMs for two periods: after pruning in the autumn and winter, and immediately before the harvest of the first flush tea leaves. However, it is not limited to this. For example, in regions prone to cold damage in winter or tea plantations frequently affected by frost damage in spring, pruning may be performed in the spring. In this case, the three-dimensional information acquisition unit 30 may acquire DSMs for two periods: after spring pruning and immediately before the harvest of the first flush tea leaves. Furthermore, in order to prevent a decline in the quality of the second flush tea leaves, pruning may be performed after a predetermined period (for example, around 10 days) has elapsed since the first flush tea leaves were picked. In such cases, the three-dimensional information acquisition unit 30 may acquire DSMs for two periods: immediately after pruning following the first flush tea leaf picking, and immediately before the harvest of the second flush tea leaves. Similarly, if pruning is performed after the second harvest, the three-dimensional information acquisition unit 30 may acquire DSM data for two periods: immediately after pruning following the second harvest, and immediately before the third harvest. Also, when the first harvest is mechanically picked, even if pruning is not performed afterward, mechanical picking ensures a uniform surface on the tea plant. For this reason, the three-dimensional information acquisition unit 30 may acquire DSM data for two periods: immediately after mechanically picking the first harvest, and immediately before the harvest of the next new shoots (second harvest, etc.). In either case, the growth rate estimation unit 32 can accurately estimate the growth rate of the new shoots between the two periods. Furthermore, the yield estimation unit 36 ​​can accurately estimate the yield based on the estimated growth rate of the new shoots between the two periods.

[0038] In the above embodiment, the case in which the server 10 performs the processing shown in Figure 4 was described, but the process is not limited to this, and the user terminal 70 may also perform the processing shown in Figure 4. In this case, the user terminal 70 should be equipped with the functions of the server 10 shown in Figure 3. Alternatively, the server 10 may have some of the functions of the user terminal 70 shown in Figure 3 (three-dimensional information generation unit 22). By not giving the user terminal 70 the functions of the three-dimensional information generation unit 22, the processing load on the user terminal 70 can be reduced.

[0039] In the above embodiment, the case in which images are taken from above using a camera 62 installed on a drone 60 (unmanned aerial vehicle) has been described, but the invention is not limited to this, and images may also be taken from a manned aircraft.

[0040] The above processing functions can be implemented by a computer. In this case, a program describing the processing content of the functions that the processing unit should have is provided. By executing this program on a computer, the above processing functions are implemented on the computer. The program describing the processing content can be recorded on a storage medium that can be read by a computer (except for carrier waves).

[0041] When distributing a program, it may be sold in the form of a portable storage medium such as a DVD (Digital Versatile Disc) or CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. Alternatively, the program can be stored in the storage device of a server computer and transferred from the server computer to other computers via a network.

[0042] A computer executing a program stores programs, for example, those recorded on a portable storage medium or transferred from a server computer, in its own memory. The computer then reads the program from its memory and executes the processing according to the program. Alternatively, the computer can directly read the program from the portable storage medium and execute the processing according to that program. Furthermore, the computer can sequentially execute the processing according to the programs received as they are transferred from the server computer.

[0043] The embodiments described above are preferred examples of the present invention. However, the invention is not limited thereto, and various modifications are possible without departing from the spirit of the invention. [Explanation of Symbols]

[0044] 10 servers 20 Image and location information acquisition unit 22 Three-dimensional information generation section 24 Three-dimensional information transmission unit 30 3D information acquisition section 32 Growth amount estimation section 34. Unit for acquiring information on pruning height 36 Yield Estimation Section 60 Drones 62 Cameras 64 GNSS sensors 70 User terminals 80 Networks 100 Tea Tree Information Estimation System

Claims

1. First three-dimensional information showing the distribution of height information of the tea trees during a first period in which the state after pruning or after mechanical harvesting of tea leaves is maintained, and second three-dimensional information showing the distribution of height information of the tea trees during a second period following the first period are obtained. Based on the difference between the second three-dimensional information and the first three-dimensional information, the amount of growth of new shoots that have grown in the tea plant after the first period is estimated. A method for estimating tea tree information, characterized in that the processing is performed by a computer.

2. We obtain information on the height at which the tea leaves are picked. Based on the growth rate of the new shoots and the information on the height at which the tea leaves are picked, the yield of the tea leaves is estimated. The method for estimating tea tree information according to claim 1, characterized in that the processing is performed by the computer.

3. The method for estimating tea tree information according to claim 1 or 2, characterized in that the first three-dimensional information and the second three-dimensional information are numerical surface models obtained from aerial images.

4. The method for estimating tea tree information according to claim 1 or 2, characterized in that the first three-dimensional information and the second three-dimensional information are information obtained by laser measurement from above.

5. The first period is the period from after the tea plants have been pruned or after the tea leaves have been mechanically picked until new shoots sprout. The second period is the period after the new shoots have sprouted. The method for estimating tea tree information according to claim 1 or 2.

6. First three-dimensional information showing the distribution of height information of the tea trees during a first period in which the state after pruning or after mechanical harvesting of tea leaves is maintained, and second three-dimensional information showing the distribution of height information of the tea trees during a second period following the first period are obtained. Based on the difference between the second three-dimensional information and the first three-dimensional information, the amount of growth of new shoots that have grown in the tea plant after the first period is estimated. A tea tree information estimation program characterized by having a computer perform the processing.

Images