Estimation device, estimation method, estimation program, recording device and recording medium

Abstract

To provide an estimation device, an estimation method, an estimation program, a recording device and a recording medium, capable of increasing the scope of utilization of vegetation indices.SOLUTION: An information processing device 20 includes: an acquisition unit 211 for acquiring a vegetation index based on images of plants growing in a predetermined area, captured across a plurality of wavelength bands, and a plant coverage ratio; and a plant height estimation unit 212 that estimates the height of plants based on the acquired vegetation index and coverage ratio.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to an estimation device, an estimation method, an estimation program, a recording device, and a recording medium. [Background technology]

[0002] In agriculture, in order to grow high-quality, stable, high-yield crops and other plants, appropriate management of the timing and amount of top dressing, measures to reduce lodging, etc. is required. Plant producers manage plants while checking the degree of growth. Plant producers check the degree of growth, for example, by measuring the height of the plant.

[0003] Meanwhile, a method has been proposed in which a vegetation index such as NDVI (Normalized Difference Vegetation Index) is obtained by photographing a farm field using a camera (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-56236 Summary of the Invention [Problem to be solved by the invention]

[0005] It would be preferable for such vegetation indices to be used more widely.

[0006] The present invention has been made in consideration of the above-mentioned circumstances, and an object of the present invention is to provide an estimation device, an estimation method, an estimation program, a recording device, and a recording medium that enable more widespread use of vegetation indices. [Means for solving the problem]

[0007] The above object of the present invention can be achieved as follows.

[0008] (1) An estimation device comprising: an acquisition unit that acquires a vegetation index based on images of plants growing in a specified area photographed in multiple wavelength bands and a vegetation cover ratio of the plants; and a grass height estimation unit that estimates the grass height of the plants based on the acquired vegetation index and vegetation cover ratio.

[0009] (2) The estimation device described in (1) above, further including a leaf color estimation unit that estimates a leaf color of the plant based on the estimated plant height.

[0010] (3) The estimation device according to (1) or (2) above, further comprising an output unit that outputs information relating to the estimated plant height.

[0011] (4) The estimation device according to any one of (1) to (3) above, wherein the grass height estimation unit estimates the grass height based on a relational equation between the vegetation index, the vegetation coverage rate, and the grass height that is stored in advance.

[0012] (5) An estimation device described in any of (1) to (4) above, further comprising a calculation unit that calculates a scattering index representing the scattering state of the plant based on the image, and the grass height estimation unit estimates the grass height based on the calculated scattering index and the acquired vegetation index and vegetation cover ratio.

[0013] (6) The estimation device according to (5) above, wherein the acquired vegetation index is an average of pixel vegetation indices for each unit pixel of the image, and the calculation unit calculates the scattering index using a histogram of the pixel vegetation index.

[0014] (7) The estimation device according to (6) above, wherein the calculation unit calculates the scattering index using a mean value and a mode of the histogram.

[0015] (8) The estimation device according to (6) or (7) above, wherein the histogram is created using the pixel vegetation index of the unit pixel in which the plant is present in the image.

[0016] (9) The estimation device according to any one of (1) to (8) above, wherein the plurality of wavelength bands include a visible light region and an infrared light region.

[0017] (10) The estimation device according to any one of (1) to (9) above, wherein the acquired vegetation coverage ratio is calculated based on the image.

[0018] (11) An estimation method including the steps of: (a) acquiring a vegetation index based on images of plants growing in a predetermined area photographed in a plurality of wavelength bands and a vegetation cover ratio of the plants; and (b) estimating a height of the plants based on the acquired vegetation index and the vegetation cover ratio.

[0019] (12) An estimation program for causing a computer to execute the estimation method according to (11) above.

[0020] (13) A recording device comprising: an acquisition unit that acquires a vegetation index based on images of plants growing in a specified area photographed in multiple wavelength bands, a vegetation cover ratio of the plants, and a plant height of the plants; a determination unit that determines a relational equation between the acquired vegetation index, the vegetation cover ratio, and the plant height; and a recording unit that records information regarding the determined relational equation on a recording medium.

[0021] (14) A recording medium on which information regarding the relational expression is recorded by the recording device described in (13) above. Effect of the Invention

[0022] According to the estimation device, estimation method, and estimation program of the present invention, the vegetation cover rate of a plant is acquired together with the vegetation index, and the plant height of the plant is estimated based on the vegetation index and the vegetation cover rate. In other words, the vegetation index is converted to the plant height using the vegetation cover rate. This allows plant producers, etc., who are unfamiliar with handling vegetation indices, to easily grasp the degree of plant growth by utilizing the vegetation index. This makes it possible to more widely utilize the vegetation index. The recording device and recording medium of the present invention record the relational equation between the vegetation index, the vegetation cover rate, and the plant height. [Brief description of the drawings]

[0023] [Figure 1] FIG. 1 is a block diagram illustrating an example of the configuration of a plant height estimation system according to a first embodiment. [Diagram 2] 2 is a diagram illustrating an example of a state in which the light measurement device illustrated in FIG. 1 is used. FIG. [Diagram 3] 2 is a block diagram illustrating an example of a configuration of the information processing device illustrated in FIG. 1. [Figure 4] 2 is a block diagram illustrating an example of a functional configuration of the information processing device illustrated in FIG. 1. [Diagram 5] 2A to 2D are diagrams illustrating examples of images captured by the light measurement device shown in FIG. [Figure 6] FIG. 5 is a diagram illustrating an example of a relational expression used in the plant height estimation unit illustrated in FIG. 4. [Figure 7] 2 is a flowchart illustrating an example of a process executed by the information processing device illustrated in FIG. 1. [Figure 8] FIG. 1 is a diagram showing the relationship between NDVI and growth volume of a given farm field. [Figure 9] This is a graph in which the vertical axis of the graph shown in FIG. 8 is converted into the product of plant height and vegetation coverage. [Figure 10] FIG. 11 is a block diagram illustrating an example of a functional configuration of an information processing device according to a first modified example. [Figure 11] FIG. 11 is a diagram illustrating an example of a relational expression used in the leaf color estimation unit shown in FIG. 10. [Figure 12] FIG. 1 shows the relationship between NDVI for a given field and the product of leaf color, plant height, and vegetation coverage. [Figure 13] 11 is a flowchart illustrating an example of a process executed by the information processing device illustrated in FIG. [Figure 14] FIG. 11 is a block diagram illustrating an example of a functional configuration of an information processing device according to Modification 2. [Figure 15] (A) and (B) are histograms of pixel NDVI at points A and B shown in Figure 9, respectively. [Figure 16] 15 is a diagram illustrating an example of a scattering index calculated by a calculation unit illustrated in FIG. 14. [Figure 17] FIG. 17 is a diagram illustrating the relationship between the scattering index and the vegetation index shown in FIG. 16. [Figure 18] FIG. 15 is a diagram illustrating an example of a relational expression used by a plant height estimation unit illustrated in FIG. 14. [Figure 19] 15 is a flowchart illustrating an example of a process executed by the information processing device illustrated in FIG. 14. [Figure 20] 20 is a flowchart illustrating another example of the process shown in FIG. 19. [Figure 21] FIG. 11 is a block diagram illustrating an example of the configuration of a recording apparatus according to a second embodiment. [Figure 22] 22 is a block diagram illustrating an example of a functional configuration of the recording device illustrated in FIG. 21. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] Hereinafter, an embodiment of an estimation device, an estimation method, an estimation program, a recording device, and a recording medium of the present invention will be described with reference to the attached drawings. Note that the same reference numerals are used for the same components in the drawings. Also, the dimensional ratios of the drawings are exaggerated for the convenience of explanation, and may differ from the actual ratios.

[0025] [First embodiment] <Overall configuration of plant height estimation system 1> Fig. 1 shows an example of the configuration of a plant height estimation system 1 according to a first embodiment of the present invention. The plant height estimation system 1 estimates the plant height of a plant (for example, a plant P in Fig. 2 described below) by so-called remote sensing. This plant height estimation system 1 includes, for example, a light measurement device 10 and an information processing device 20. The light measurement device 10 and the information processing device 20 are connected, for example, via a network. Here, the information processing device 20 corresponds to a specific example of an estimation device of the present invention.

[0026] The light measurement device 10 includes a camera such as a multispectral camera or a hyperspectral camera, and captures images of plants in a field in a plurality of wavelength bands under the control of the information processing device 20 to generate images. The light measurement device 10 includes, for example, a visible light measurement unit 10A and an infrared light measurement unit 10B, and generates images of visible light (for example, wavelengths of 400 nm to 700 nm) and images of infrared light (for example, wavelengths of 800 nm to 1 mm). The visible light measurement unit 10A generates images in, for example, a red wavelength region (near 610 nm), a green wavelength region (near 550 nm), or a blue wavelength region (near 470 nm). The infrared light measurement unit 10B generates images in, for example, a near-infrared wavelength region (800 nm to 1300 nm). Image data of the plants captured by the light measurement device 10 is sent to the information processing device 20.

[0027] The light measurement device 10 includes, for example, a bandpass filter, an imaging optical system, an image sensor, and a digital signal processor (DSP). The bandpass filter selectively transmits light in a predetermined wavelength range (visible light range and infrared light range). The imaging optical system forms an optical image of the light transmitted through the bandpass filter on a predetermined imaging plane. The image sensor has a light receiving surface at the same position as the imaging plane of the imaging optical system, and converts the optical image in the predetermined wavelength range into an electrical signal. The digital signal processor processes the electrical signal output from the image sensor to generate image data.

[0028] FIG. 2 shows an example of a situation where a plant P in a field is photographed using the light measurement device 10. The light measurement device 10 is mounted on an unmanned aerial vehicle such as a drone, and photographs the plant P in the field from the sky. By photographing the plant P in the field from the sky in this way, it is possible to easily photograph the entire field in a short time. The plant P is, for example, a crop such as rice. When sunlight L is irradiated onto the leaves of the plant P, the sunlight L is reflected (reflected light Lr), scattered (scattered light Ls), and transmitted (transmitted light Lt) by the leaves of the plant P. The light measurement device 10 is mounted on a drone or the like so as to receive the reflected light Lr and scattered light Ls from the leaves of the plant P. The light measurement device 10 can detect changes such as multiple reflection and multiple scattering according to the state of the leaves of the plant P.

[0029] The information processing device 20 is, for example, a computer such as a PC (Personal Computer) or a tablet terminal. The information processing device 20, for example, controls the imaging of the plant P by the light measurement device 10, and estimates the plant height of the plant P based on the image data of the plant P captured by the light measurement device 10.

[0030] 3 is a block diagram showing an example of a schematic configuration of the information processing device 20. The information processing device 20 has, for example, a CPU (Central Processing Unit) 21, a ROM (Read Only Memory) 22, a RAM (Random Access Memory) 23, a storage 24, a communication interface 25, and an operation display unit 26. Each component is connected to each other via a bus 27 so as to be able to communicate with each other.

[0031] The CPU 21 controls the above components and performs various arithmetic processing in accordance with programs recorded in the ROM 22 and the storage 24. The specific functions of the CPU 21 will be described later.

[0032] The ROM 22 stores various programs and various data.

[0033] The RAM 23 serves as a working area for temporarily storing programs and data.

[0034] The storage 24 stores various programs including an operating system and various data. The storage 24 may store a relationship between a vegetation index and a plant height of the plant P described later, or may have installed therein an application for estimating the plant height of the plant P from image data of the plant P using a trained classifier. The storage 24 may also store information about an image of the plant P captured by the light measurement device 10. The storage 24 may store a trained model used as a classifier and teacher data used in machine learning.

[0035] The communication interface 25 is an interface for communicating with other devices. A communication interface according to various wired or wireless standards is used as the communication interface 25. The communication interface 25 is used, for example, when receiving image data of the plant P from the light measurement device 10 or an external device, or when transmitting shooting conditions to the light measurement device 10.

[0036] The operation display unit 26 is configured, for example, by a touch panel including a display unit such as an LCD (liquid crystal display) or an organic EL display, and a touch sensor. The display unit that displays various information and an operation unit that accepts various operations by the user may be provided separately. In this case, the display unit may be configured by the above-mentioned display, viewer software, a printer, etc., and the operation unit may be configured by a touch sensor and a pointing device such as a mouse, a keyboard, etc.

[0037] <Functions of the information processing device 20> 4 is a block diagram showing the functional configuration of the information processing device 20. The information processing device 20 functions as an acquisition unit 211, a grass height estimation unit 212, and an output unit 213 by the CPU 21 reading a program stored in the storage 24 and executing processing.

[0038] The acquisition unit 211 acquires the vegetation index and vegetation coverage rate of the plant P at a given time point. The vegetation index is an index devised for the purpose of grasping the state of vegetation with a simple calculation formula using satellite data or the like by utilizing the characteristics of light reflection by the plant P, and indicates the amount and vitality of the plant P. This vegetation index is calculated based on images of the plant P captured in a plurality of wavelength bands, specifically, images captured by the light measurement device 10. Examples of the vegetation index include NDVI and GNDVI. NDVI is expressed, for example, by the following formula (1), and GNDVI is expressed, for example, by the following formula (2).

[0039]

number

[0040] The vegetation index acquired by the acquisition unit 211 is preferably GNDVI. Since green light has a higher reflectance of plants P than red light, noise can be suppressed. In addition, the difference between the reflectance of green light and that of infrared light (R IR -R GREEN ) is less likely to saturate. Therefore, GNDVI can detect vegetation activity with higher accuracy than NDVI.

[0041] The acquisition unit 211 may acquire vegetation indices other than NDVI and GNDVI, such as modified NDVI (mNDVI), Simple Ratio (SR), Specific Growth Rate (SGR), Photochemical Reflectance Index (PRI), Relative Growth Rate (RGR), Normalized Pigment Chlorophyll Index (NPCI), Simple Ratio Pigment Index (SRPI), Normalized Phaeophytinization Index (NPQI), Structure-Insensitive Pigment Index (SIPI), PI1, PI2, PI3, and PI4.

[0042] The information processing device 20 calculates a vegetation index (hereinafter referred to as pixel vegetation index) for each unit pixel of an image captured by the light measurement device 10, for example. The vegetation index acquired by the acquisition unit 211 is, for example, an average of these pixel vegetation indices. The information processing device 20 calculates a vegetation index (average of pixel vegetation indices) based on the pixel vegetation index of a unit pixel in which it is determined that a plant P exists, among the images captured by the light measurement device 10. The information processing device 20 determines whether or not a plant P exists in the unit pixel, for example, based on a threshold value of the pixel vegetation index. The threshold value of this vegetation index can be set arbitrarily, for example.

[0043] The vegetation coverage rate of the plants P is the ratio of the area of ​​the ground surface that is occupied by vegetation per unit area. The vegetation coverage rate of the plants P is preferably calculated based on an image captured by the light measurement device 10. This makes it possible to calculate both the vegetation index and the vegetation coverage rate based on the image captured by the light measurement device 10. In addition, by using the image, the vegetation coverage rate of the plants P in the entire field can be easily obtained, making it easier to estimate the plant height of the plants P in the entire field.

[0044] 5(A) to (D) show examples of images of the foliage of the plant P in the field photographed by the light measurement device 10. FIG. 5(A) shows an image of the plant P 30 days after transplantation, FIG. 5(B) shows an image of the plant P 35 days after transplantation, FIG. 5(C) shows an image of the plant P 37 days after transplantation, and FIG. 5(D) shows an image of the plant P 43 days after transplantation. Based on these images, for example, the information processing device 20 calculates the NDVI and vegetation coverage ratio on each of the 30th, 35th, 37th, and 43rd days after transplantation. For example, the NDVI on the 30th day after transplantation is 0.30, and the vegetation coverage ratio is 0.44, the NDVI on the 35th day after transplantation is 0.44, and the vegetation coverage ratio is 0.80, the NDVI on the 37th day after transplantation is 0.55, and the vegetation coverage ratio is 0.89, and the NDVI on the 43rd day after transplantation is 0.66, and the vegetation coverage ratio is 0.93. The acquisition unit 211 acquires, for example, values ​​of NDVI and vegetation coverage calculated from such an image.

[0045] The plant height estimation unit 212 estimates the plant height of the plant P based on the vegetation index and vegetation coverage rate of the plant P at a predetermined time point acquired by the acquisition unit 211. The plant height is the height of the above-ground part of the plant, that is, the height from the ground to the tip. For example, the producer of the plant P determines the growth stage of the plant P based on this plant height, and manages the timing and amount of top dressing, measures to reduce lodging, and the like. In this embodiment, the plant height estimation unit 212 estimates the plant height of the plant P based on the vegetation index and vegetation coverage rate of the plant P at a predetermined time point. This allows the vegetation index to be used more widely, as will be described in detail later.

[0046] The grass height estimation unit 212 estimates the grass height of the plant P at a predetermined time based on the relationship between the vegetation index, vegetation coverage rate, and grass height stored in the storage 24, for example. The storage 24 stores, for example, a relational expression between the vegetation index, vegetation coverage rate, and grass height of the plant P that has been obtained in advance. The relational expression between the vegetation index, vegetation coverage rate, and grass height of the plant P is obtained, for example, using data of the past plant P. The relational expression between the vegetation index, vegetation coverage rate, and grass height of the plant P may be obtained using data from the past few years. The relational expression between the vegetation index, vegetation coverage rate, and grass height of the plant P may be updated every predetermined period. At this time, the grass height of the plant P is measured using a measuring instrument such as a ruler. The grass height may be measured by mounting a LiDAR (Light Detection and Ranging) or the like on a drone, or may be measured using parallax from an orthoimage. The orthoimage is generated, for example, from a plurality of images of the plant P taken by a camera mounted on the drone.

[0047] Fig. 6 shows an example of the relationship between the vegetation index, vegetation coverage rate, and plant height of a plant P. The vertical axis of Fig. 6 represents the product of the plant height and the vegetation coverage rate (plant height x vegetation coverage rate), and the horizontal axis represents the vegetation index. The relationship between the vegetation index, vegetation coverage rate, and plant height of a plant P is expressed by a linear function, for example, as shown in the following formula (3).

[0048]

number

[0049] The plant height estimation unit 212 estimates the plant height of the plant P at a predetermined time point by using, for example, the vegetation index and the vegetation coverage rate acquired by the acquisition unit 211 and equation (3). The relational equation between the vegetation index, the vegetation coverage rate, and the plant height of the plant P is not limited to a linear correlation.

[0050] The plant height estimation unit 212 may estimate the plant height of the plant P by utilizing machine learning. For example, the plant height estimation unit 212 uses a trained classifier to input the vegetation index and vegetation coverage rate of the plant P acquired by the acquisition unit 211, and estimates the plant height of the plant P. The trained classifier is stored, for example, in the storage 24. In this trained classifier, the vegetation index and vegetation coverage rate of the plant P calculated from past images, and the measured value of the plant height are trained in advance as training data.

[0051] The output unit 213 outputs information about the height of the plant P at a predetermined time point estimated by the grass height estimation unit 212. The information about the grass height is, for example, information indicating the value of the grass height, information indicating whether the grass height is greater than or less than a predetermined grass height, and information indicating the amount of excess or deficiency relative to the predetermined grass height. The output unit 213 outputs information about the grass height, for example, by displaying the value of the grass height on the operation display unit 26. The output unit 213 may display information such as "Poor growth", "Growing well", and "Excessive growth" on the operation display unit 26, or may output information about the grass height by voice.

[0052] <Processing Overview of Information Processing Device 20> The process executed in the information processing device 20, that is, the estimation method executed by the information processing device 20, will be described in detail below.

[0053] Fig. 7 is a flowchart showing the procedure of the estimation process executed in the information processing device 20. The process of the information processing device 20 shown in the flowchart of Fig. 7 is stored as a program in the storage 24 of the information processing device 20, and is executed by the CPU 21 controlling each unit.

[0054] (Step S101) The information processing device 20 first acquires the vegetation index and vegetation coverage ratio of the plant P at a predetermined time. The information processing device 20 acquires the GNDVI and vegetation coverage ratio calculated from an image of the plant P in the entire field at a predetermined time captured by the light measurement device 10, for example.

[0055] (Step S102) The information processing device 20 inputs the vegetation index and vegetation cover ratio acquired in the processing of step S101 into a relational equation between the vegetation index, vegetation cover ratio and plant height (for example, the above equation (3)) previously stored in the storage 24, and estimates the plant height of the plant P at a specified point in time.

[0056] (Step S103) The information processing device 20 outputs information about the height of the plant P at the predetermined time point estimated in the processing of step S102, and ends the processing. For example, the information processing device 20 outputs information about the height by displaying the value of the height of the plant P at the predetermined time point estimated in the processing of step S102 on the operation display unit 26.

[0057] <Functions and Effects of the Information Processing Device 20 and the Plant Height Estimation System 1> In the information processing device 20 and plant height estimation system 1 of this embodiment, the plant height of the plant P at a given time point is estimated by using the vegetation coverage ratio together with the vegetation index calculated based on the image captured by the light measurement device 10. In other words, the vegetation index is converted to plant height using the vegetation coverage ratio. As a result, even if the producer of the plant P is unfamiliar with handling vegetation indices, the vegetation index can be used to easily grasp the degree of growth of the plant P. The effects of this will be described in detail below.

[0058] Vegetation indices such as NDVI and GNDVI can be calculated from images, so they can be measured non-contact and over a wide area of ​​a field at once. This allows plant producers to easily understand the distribution of vegetation activity over the entire field from the vegetation indices. In other words, vegetation indices are useful indices that indicate the activity of vegetation.

[0059] On the other hand, plant producers have long used plant height as an index to indicate the degree of plant growth, and many producers are unfamiliar with handling vegetation indices. Plant height is measured on the ground using measuring instruments such as rulers, which is time-consuming and makes it difficult to accurately grasp the distribution of plant height throughout the entire field. Non-contact plant height measurement methods using stereo vision or laser radar have also been considered, but have not yet been put to practical use. This is because it is difficult to measure the plant height of plants that form communities, such as rice, without contact throughout the entire field. If vegetation indices could be linked to plant height, plant producers would be able to more easily use the useful vegetation indices as described above.

[0060] Figure 8 shows the relationship between NDVI and growth volume for each day after transplanting of rice (16, 23, 30, 35, 37, 43, 51, and 58 days after transplanting). The horizontal axis of Figure 8 shows NDVI, and the vertical axis shows growth volume. The growth volume of rice is determined using the following formula (4). In formula (4), the leaf color (SPAD value), stalk number, and plant height of the plant are measured on the ground. The four rice varieties are Akitakomachi, Hitomebore, Koshihikari, and Kinuhikari.

[0061]

number

[0062] From FIG. 8, it can be seen that the relationship between NDVI and growth amount shifts depending on the number of days after transplantation, in other words, depending on the growth stage. That is, it can be seen that the relationship between NDVI and growth amount shifts depending on the growth stage. In detail, in the early growth stage, the relationship between NDVI and growth amount is maintained constant, but the relationship between NDVI and growth amount starts to change about 10 days before the panicle formation stage begins. For example, it is possible to convert the vegetation index into the growth amount by determining the relationship between the vegetation index and the growth amount for each plant variety and each growth stage in advance, but this requires a huge and complicated operation. Therefore, it is difficult to put it into practical use.

[0063] In contrast to this, the plant height estimation system 1 and the information processing device 20 use the vegetation coverage ratio together with a vegetation index such as NDVI, so that it is possible to easily convert the vegetation index into plant height.

[0064] Figure 9 shows the data shown in Figure 8 re-plotted using the vegetation coverage ratio. The horizontal axis of Figure 9 shows NDVI, and the vertical axis shows the product of plant height and vegetation coverage ratio (plant height x vegetation coverage ratio). In Figure 9, the coefficient of determination R 2 The value of is 0.9655. In other words, a certain correlation is maintained among NDVI, vegetation coverage, and plant height throughout the entire growth stage, and it can be seen that vegetation indices such as NDVI have a correlation with plant volume information (plant height x vegetation coverage). Therefore, regardless of the growth stage, a single relational formula can be used to convert vegetation indices to plant height.

[0065] As described above, in the plant height estimation system 1 and the information processing device 20, the vegetation cover rate of the plant P is acquired along with the vegetation index of the plant P, so that the vegetation index can be easily converted into the plant height. As a result, even if a plant producer or the like is unfamiliar with handling vegetation indices, the vegetation index can be used to easily grasp the degree of plant growth. This allows the vegetation index to be used more widely.

[0066] In addition, because the vegetation index is linked to plant height, the vegetation index becomes familiar to plant producers and the like, making it possible to popularize the vegetation index in agriculture. This makes it easier to incorporate the vegetation index into cultivation guidelines created by agricultural experiment stations, for example. This will therefore promote the use of remote sensing in the agricultural field, and promote the realization of so-called smart agriculture.

[0067] Below, we will explain modifications and other embodiments of the information processing device 20 described in the above embodiment. In order to avoid duplication of explanation, detailed explanations of configurations similar to those described in the above embodiment will be omitted.

[0068] [Variation 1] 10 shows a functional configuration of the CPU 21 of the information processing device 20 according to Modification 1. The CPU 21 may function as a leaf color estimation unit 214 in addition to the acquisition unit 211, the plant height estimation unit 212, and the output unit 213.

[0069] The leaf color estimation unit 214 estimates the leaf color (e.g., SPAD value) of the plant P based on the relationship between the plant height of the plant P at a predetermined time point estimated by the plant height estimation unit 212 and the vegetation index and vegetation coverage ratio acquired by the acquisition unit 211. The leaf color estimation unit 214 estimates the leaf color of the plant P based on the relationship between the NDVI, vegetation coverage ratio, plant height, and leaf color stored in the storage 24, for example.

[0070] Fig. 11 shows an example of the relationship between the vegetation index, vegetation coverage rate, plant height, and leaf color (SPAD value) of a plant P. The vertical axis of Fig. 11 shows the product of the leaf color, plant height, and vegetation coverage rate (SPAD value x plant height x vegetation coverage rate), and the horizontal axis shows the vegetation index. The relationship between the vegetation index, vegetation coverage rate, plant height, and leaf color of a plant P is expressed by a linear function such as the following equation (5).

[0071]

number

[0072] Equation (5) is obtained, for example, using past data of plant P. The relational expressions among the vegetation index, vegetation coverage rate, plant height, and leaf color of plant P may be obtained using data from the past few years. The relational expressions among the vegetation index, vegetation coverage rate, plant height, and leaf color of plant P may be updated at predetermined intervals. At this time, the plant height and leaf color of plant P are measured using a measuring device.

[0073] The leaf color estimation unit 214 estimates the leaf color of the plant P at a predetermined time point using, for example, the plant height of the plant P at a predetermined time point estimated by the plant height estimation unit 212, the vegetation index and vegetation coverage rate acquired by the acquisition unit 211, and equation (5). The relational equation among the vegetation index, vegetation coverage rate, plant height, and leaf color of the plant P is not limited to a linear correlation.

[0074] Figure 12 shows the relationship between the vegetation index, vegetation coverage rate, plant height, and leaf color of rice in a given year. The vegetation index and vegetation coverage rate were measured by remote sensing, while the plant height and leaf color were measured using a measuring device on the ground. The horizontal axis of Figure 12 shows NDVI, and the vertical axis shows the product of leaf color (SPAD value), plant height, and vegetation coverage rate (SPAD value x plant height x vegetation coverage rate). In Figure 12, the coefficient of determination R 2 The value of NDVI is 0.9159. In other words, it can be seen that a certain correlation is maintained among NDVI, vegetation coverage, plant height and leaf color throughout the entire growth stage. Therefore, a single relational equation can be used to convert vegetation indices into leaf color regardless of the growth stage.

[0075] The leaf color estimation unit 214 may estimate the plant height of the plant P by utilizing machine learning. For example, the leaf color estimation unit 214 uses a trained classifier to estimate the leaf color of the plant P by inputting the vegetation index and vegetation coverage rate of the plant P acquired by the acquisition unit 211 and the plant height of the plant P estimated by the plant height estimation unit 212. The trained classifier is stored, for example, in the storage 24. In this trained classifier, the vegetation index and vegetation coverage rate of the plant P calculated from past images, and the measured values ​​of the plant height and leaf color are trained in advance as training data.

[0076] The output unit 213 outputs, for example, information on the leaf color of the plant P estimated by the leaf color estimation unit 214. The output unit 213 may output information on the plant height of the plant P estimated by the plant height estimation unit 212 together with the information on the leaf color.

[0077] FIG. 13 is a flowchart showing the procedure of the estimation process executed in the information processing device 20 according to the first modification.

[0078] (Steps S201 and S202) First, the information processing device 20 acquires the vegetation index and vegetation coverage rate of the plant P at a predetermined time point, in the same manner as in steps S101 and S102 (FIG. 7) described above, and then estimates the plant height of the plant P.

[0079] (Steps S203 and S204) Next, the information processing device 20 estimates the leaf color of the plant P based on the plant height of the plant P estimated in the processing of step S202. The information processing device 20 estimates the leaf color of the plant P, for example, by inputting the plant height of the plant P estimated in the processing of step S202 and the vegetation index and vegetation coverage rate of the plant P acquired in the processing of step S201 into the above formula (5). Thereafter, the information processing device 20 outputs information on the leaf color of the plant P estimated in the processing of step S203 and ends the processing. The information processing device 20 may output information on the plant height estimated in step S202 together with information on the leaf color of the plant P.

[0080] The information processing device 20 having such a CPU 21, like that described in the above embodiment, can easily convert the vegetation index into plant height because the vegetation coverage rate of the plant P is obtained together with the vegetation index of the plant P. As a result, even if a plant producer or the like is unfamiliar with handling vegetation indices, the vegetation index calculated from the image can be used to easily grasp the growth level of the plant. This makes it possible to more widely utilize the vegetation index. Furthermore, in this information processing device 20, the leaf color of the plant P is estimated, so that the producer or the like of the plant P can obtain information that is further useful for the growth of the plant.

[0081] [Variation 2] 14 shows a functional configuration of the CPU 21 of the information processing device 20 according to Modification 2. The CPU 21 may function as a calculation unit 215 in addition to the acquisition unit 211, the plant height estimation unit 212, and the output unit 213.

[0082] The calculation unit 215 calculates a scattering index representing a scattering state of the plant P based on an image of the plant P photographed by the light measurement device 10. The calculation unit 215 calculates the scattering index using, for example, a histogram of pixel vegetation indices calculated for each unit pixel of the plant P photographed by the light measurement device 10.

[0083] Fig. 15(A) shows a histogram of the pixel vegetation index of point A (NDVI is 0.49) shown in Fig. 9, and Fig. 15(B) shows a histogram of the pixel vegetation index of point B (NDVI is 0.65) shown in Fig. 9. Specifically, the vertical axis of Fig. 15(A) and (B) shows the number of pixels, and the horizontal axis shows the NDVI of each unit pixel (hereinafter referred to as pixel NDVI). The NDVI values ​​of points A and B shown in Fig. 9, 0.49 and 0.65, are the average values ​​of the respective pixel NDVIs.

[0084] The most frequent value of the pixel NDVI shown in Figure 15(A) is around 0.60, and the tail is long in the area of ​​values ​​smaller than this (left side). This indicates that in the case of plant P in the state of point A, the degree of overlap of the leaves is relatively small, and the degree of multiple reflection, i.e., the degree of scattering, is small.

[0085] On the other hand, the most frequent value of the pixel NDVI shown in Fig. 15(B) is around 0.70, and the symmetry of the regions to the left and right of the most frequent value is higher than that in Fig. 15(A). This indicates that the plant P in the state of point B is closer to a foliage with uniform overlapping leaves, and has a higher degree of multiple reflection, i.e., scattering, than the plant P in the state of point A.

[0086] FIG. 16 shows an example of a method for calculating a scattering index that indicates the degree of scattering. FIG. 16 shows a histogram of the pixel vegetation index of a plant P, with the vertical axis showing the number of pixels and the horizontal axis showing the pixel NDVI. The calculation unit 215 calculates, for example, the difference ΔNDVI between the mode Mo and the average value Me of this histogram (ΔNDVI=|mode Mo-average value Me|) as the scattering index. When ΔNDVI is small, it indicates that the degree of scattering of the plant P is relatively large, and when ΔNDVI is large, it indicates that the degree of scattering of the plant P is relatively small. It is preferable that such a histogram is created using the pixel vegetation index of the unit pixel where the plant P is present in the image captured by the light measurement device 10. This makes it possible to calculate the scattering index with higher accuracy.

[0087] FIG. 17 shows an example of the relationship between the vegetation index and the scattering index. The vertical axis of FIG. 17 is ΔNDVI, and the horizontal axis is NDVI (average of pixel NDVI). For example, when the NDVI value is small, the overlap of the leaves of the plant P is small, so that ΔNDVI is large. The storage 24 stores, for example, a relational expression between the vegetation index and scattering index of the plant P. The relational expression between the vegetation index and scattering index of the plant P is obtained, for example, by using data of the plant P in the past. For example, the relational expression between the vegetation index, vegetation coverage rate, and plant height of the plant P shown in the above-mentioned formula (3) can be obtained at the same time.

[0088] The grass height estimation unit 212 estimates the grass height of the plant P at a predetermined time point based on the scattering index calculated by the calculation unit 215 and the vegetation index and vegetation coverage rate acquired by the acquisition unit 211. The grass height estimation unit 212 estimates the grass height of the plant P at a predetermined time point, for example, as follows.

[0089] First, a correction index Is of the plant P at a predetermined time point is calculated using a relational expression (for example, FIG. 17) between the vegetation index and the scattering index of the plant P stored in the storage 24. The correction index Is is an index that indicates the degree of deviation between the scattering index of the plant P calculated from the relational expression stored in the storage 24 and the scattering index calculated by the calculation unit 215. For example, when the NDVI value acquired by the acquisition unit 211 is 0.5, the correction index Is (Is2) can be calculated according to the degree of deviation between the scattering index of the plant P calculated from the relational expression stored in the storage 24 (point C) and the scattering index calculated by the calculation unit 215 (point D).

[0090] Next, the plant height estimation unit 212 estimates the plant height of the plant P at a predetermined time based on, for example, the relationship between the correction index Is, the vegetation index, the vegetation coverage rate, and the plant height stored in the storage 24. The storage 24 stores, for example, a relational expression between the correction index Is, the vegetation index, the vegetation coverage rate, and the plant height of the plant P that has been obtained in advance. The relational expression between the correction index Is, the vegetation index, the vegetation coverage rate, and the plant height of the plant P has been obtained, for example, by using past data of the plant P.

[0091] Fig. 18 shows an example of the relationship between the correction index Is, the vegetation index, the vegetation coverage rate, and the plant height of a plant P. The vertical axis of Fig. 18 shows the product of the plant height and the vegetation coverage rate (plant height x vegetation coverage rate), and the horizontal axis shows the vegetation index. For example, the storage 24 stores a relational expression between the vegetation index, the vegetation coverage rate, and the plant height for each correction index Is (Is-1, Is-2, Is0, Is1, Is2).

[0092] The plant height estimation unit 212 estimates the plant height of the plant P at a predetermined time point using, for example, the relational expression shown in FIG. 18, the correction index Is (Is2) previously obtained, and the vegetation index and vegetation coverage rate acquired by the acquisition unit 211. For example, when the correction index Is2, the NDVI is 0.5, and the vegetation coverage rate is 0.75 (point D'), the plant height estimation unit 212 estimates the plant height of the plant P to be about 133 cm (100 / 0.75). Note that when the correction index is Is0, even if the NDVI is 0.5 and the vegetation coverage rate is 0.75 (point C'), the plant height estimation unit 212 estimates the plant height of the plant P to be about 87 cm (65 / 0.75). The output unit 213 outputs information on the plant height of the plant P at a predetermined time point estimated by the plant height estimation unit 212.

[0093] FIG. 19 is a flowchart showing the procedure of the estimation process executed in the information processing device 20 according to the second modification.

[0094] (Step S301) First, the information processing device 20 acquires the vegetation index and the vegetation coverage rate of the plant P at a predetermined time point in the same manner as in step S101 (FIG. 7) described above.

[0095] (Step S302) The information processing device 20 calculates the scattering index of the plant P at a predetermined time based on the image of the plant P captured by the light measurement device 10.

[0096] (Step S303) The information processing device 20 inputs the vegetation index and vegetation coverage rate obtained in the processing of step S301 and the scattering index calculated in the processing of step S302 into a relational equation (for example, Figure 18 described above) between the scattering index, vegetation index, vegetation coverage rate and grass height that is pre-stored in the storage 24, and estimates the grass height of the plant P at a specified point in time.

[0097] (Step S304) The information processing device 20 outputs information about the height of the plant P at the predetermined time point estimated in the process of step S303, and ends the process.

[0098] 20 shows another example of the procedure of the estimation process executed in the information processing device 20. As described in the above modification example 1, the information processing device 20 may further estimate a leaf color (step S305) and output information related to the leaf color (step S306).

[0099] In the information processing device 20 having such a CPU 21, as described in the above embodiment, the vegetation coverage rate of the plant P is acquired together with the vegetation index of the plant P, so that the vegetation index can be easily converted into the plant height. As a result, even if a plant producer or the like is unfamiliar with handling vegetation indices, the vegetation index calculated from the image can be used to easily grasp the growth level of the plant. This makes it possible to more widely utilize the vegetation index. In addition, in this information processing device 20, the scattering index is calculated, so that the plant height and leaf color of the plant P can be estimated with higher accuracy. This will be described below.

[0100] The vegetation coverage ratio acquired by the acquisition unit 211 is an important value that represents the distribution of the plants P throughout the entire field, but this vegetation coverage ratio does not reflect the degree of overlap of the leaves of the plants P. In other words, even if the leaves of the plants P are in a state where there is little overlap, or even if the plants P have grown to a state where the leaves of the plants P overlap more, the vegetation coverage ratio may not change significantly. In this modified example, the degree of overlap of the leaves of the plants P is calculated as a scattering index, so that the growth state of the plants P is reflected in a more multifaceted manner, and the plant height can be estimated with higher accuracy.

[0101] [Second embodiment] <Configuration of Recording Device 30> 21 is a block diagram showing an example of a schematic configuration of a recording device 30 according to a second embodiment of the present invention. The recording device 30 is, for example, a printer, and has a CPU 31, a ROM 32, a RAM 33, a storage 34, a communication interface 35, an operation display unit 36, and a recording unit 37. Each component is connected to each other via a bus 38 so as to be able to communicate with each other. The functions of the CPU 31, the ROM 32, the RAM 33, the storage 34, the communication interface 35, and the operation display unit 36 ​​are similar to the functions of the CPU 21, the ROM 22, the RAM 23, the storage 24, the communication interface 25, and the operation display unit 26 of the information processing device 20. The recording device 30 may have the functions of the information processing device 20 described in the first embodiment and the like.

[0102] The recording unit 37 forms an image on a recording medium such as paper using a known imaging process such as an electrophotographic process. The recording unit 37 records, for example, an image relating to the relational expression between the vegetation index, vegetation coverage rate, and plant height of the plant P (for example, the graph shown in FIG. 11 above) on the recording medium. The recording unit 37 may also record, for example, an image relating to the relational expression between the correction index Is, vegetation index, vegetation coverage rate, and plant height of the plant P (for example, the graph shown in FIG. 18 above) on the recording medium. The recording unit 37 may further record information relating to the vegetation index and vegetation coverage rate acquired by the information processing device 20 (for example, point D' in FIG. 18).

[0103] 22 is a block diagram showing the functional configuration of the recording device 30. The recording device 30 functions as an acquisition unit 311, a determination unit 312, and an output unit 313 by the CPU 31 reading a program stored in the storage 34 and executing the process.

[0104] The acquisition unit 311 acquires over time the vegetation index, the vegetation coverage rate, and the plant height of the plant P. For example, the acquisition unit 311 acquires the vegetation index and the vegetation coverage rate calculated based on an image captured by the light measurement device 10, and acquires the plant height measured using a measuring instrument such as a ruler.

[0105] The determining unit 312 determines a relational expression between the vegetation index, the vegetation coverage rate, and the plant height acquired by the acquiring unit 311 (for example, the above-mentioned expression (3)).

[0106] The output unit 313 outputs information about the relational expression determined by the determination unit 312 to the recording unit 37. In response to this, the recording unit 37 records an image related to this relational expression on a recording medium.

[0107] The recording device 30 and the recording medium record images and the like relating to the relational expressions between the vegetation index, vegetation coverage rate, and plant height of the plant P. This allows plant producers and the like who are unfamiliar with handling vegetation indices to easily grasp the degree of plant growth by utilizing the vegetation indices, thereby enabling the vegetation indices to be used more widely.

[0108] As described above, the estimation device, estimation method, estimation program, recording device, and recording medium of the present invention have been described in the embodiments and modifications. However, it goes without saying that those skilled in the art can appropriately add to, modify, and omit the present invention within the scope of the technical concept thereof.

[0109] For example, in the above embodiment, the plant P is rice, but the plant P may be soybean, adzuki bean, wheat, or the like.

[0110] In the above embodiment, an example has been described in which the grass height estimation system 1 includes the light measurement device 10 and the information processing device 20, but the grass height estimation system 1 may further include a vegetation index correction unit that corrects the vegetation index, etc. The vegetation index correction unit corrects the vegetation index calculated from the image according to, for example, the intensity of sunlight, the altitude, etc.

[0111] In addition, in the above embodiment and modified examples, examples have been described in which the relational equation or the trained classifier is stored in storage 24, but the relational equation and the trained classifier may be stored in an external storage device, etc.

[0112] Furthermore, the processing units in the flowcharts in the above embodiments are divided according to the main processing contents in order to facilitate understanding of each process. The method of classifying the processing steps does not limit the present invention. Each process can be further divided into more processing steps. Also, one processing step may execute more processes.

[0113] The means and methods for performing various processes in the system according to the above-mentioned embodiment can be realized by either a dedicated hardware circuit or a programmed computer. The above-mentioned program may be provided by a computer-readable recording medium such as a flexible disk or a CD-ROM, or may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording medium is usually transferred to and stored in a storage unit such as a hard disk. The above-mentioned program may be provided as a standalone application software, or may be incorporated into the software of the device as one function of the system. [Explanation of symbols]

[0114] 1. Grass height estimation system, 10 Light measuring device, 10A visible light measurement section, 10B Infrared light measurement section, 20 Information processing device, 21,31 CPU, 211 Acquisition Department; 212 Plant height estimation section, 213 output section, 214 Leaf color estimation section, 215 Calculation unit, 22,32 ROM, 23,33 RAM, 24,34 storage, 25,35 communication interface, 26,36 Operation display section, 30 Recording device, 37 Records Section.

Claims

1. an acquisition unit that acquires a vegetation index based on images of plants growing in a predetermined area captured in a plurality of wavelength bands and a vegetation coverage rate of the plants; a plant height estimation unit that estimates a plant height of the plant based on the acquired vegetation index and the vegetation coverage rate; An estimation device comprising:

2. The estimation device according to claim 1 , further comprising a leaf color estimation unit that estimates a leaf color of the plant based on the estimated plant height.

3. The estimation device according to claim 1 , further comprising an output unit that outputs information related to the estimated plant height.

4. The estimation device according to claim 1 , wherein the grass height estimation unit estimates the grass height based on a relational expression between the vegetation index, the vegetation coverage rate, and the grass height that is stored in advance.

5. A calculation unit is further provided to calculate a scattering index representing a scattering state of the plant based on the image, The estimation device according to claim 1 , wherein the grass height estimation unit estimates the grass height based on the calculated scattering index, and the acquired vegetation index and vegetation coverage rate.

6. The obtained vegetation index is an average of pixel vegetation indices for each unit pixel of the image, The estimation device according to claim 5 , wherein the calculation unit calculates the scattering index using a histogram of the pixel vegetation index.

7. The estimation device according to claim 6 , wherein the calculation unit calculates the scattering index using a mean value and a mode of the histogram.

8. The estimation device according to claim 6 , wherein the histogram is created using the pixel vegetation index of the unit pixel in which the plant is present in the image.

9. The estimation device according to claim 1 , wherein the plurality of wavelength bands include a visible light range and an infrared light range.

10. The estimation device according to claim 1 , wherein the acquired vegetation coverage ratio is calculated based on the image.

11. (a) acquiring a vegetation index based on images of plants growing in a predetermined area captured in a plurality of wavelength bands and a vegetation coverage rate of the plants; (b) estimating the height of the plant based on the acquired vegetation index and the vegetation coverage rate; An estimation method including:

12. An estimation program that causes a computer to execute the estimation method according to claim 11.

13. an acquisition unit that acquires a vegetation index based on images of plants growing in a predetermined area captured in a plurality of wavelength bands, a vegetation coverage rate of the plants, and a plant height of the plants; A determination unit that determines a relational expression between the acquired vegetation index, the vegetation coverage rate, and the plant height; a recording unit for recording information about the determined relational expression on a recording medium; A recording device comprising:

14. A recording medium on which information relating to the relational expression is recorded by the recording device according to claim 13.

Images