Information processing device and index value calculation method

Abstract

Provided is an information processing device capable of calculating accurate index values ​​related to the growth status of a plant without increasing the burden on a user. The information processing device comprises: a receiving unit that receives input of a captured image, the captured image being obtained by capturing a cultivation area of ​​a plant including a rectangular region from the length direction of the rectangular region; a transformation coefficient calculation unit that determines the number of pixels in the width direction of the rectangular region in the captured image for each of a plurality of depth direction positions in the captured image input to the receiving unit, and calculates a transformation coefficient for each depth direction position by dividing the actual width of the rectangular region by each determined number of pixels; and an index value calculation unit that calculates an index value related to the growth status of the plant by analyzing the captured image using the transformation coefficient calculated by the transformation coefficient calculation unit for each depth direction position.

Description

Technical Field

[0001] The present invention relates to an information processing device for calculating an index value related to a growth state of a plant and an index value calculation method. Background Art

[0002] There is a method for evaluating the growth state of plants by analyzing images obtained by photographing plants (for example, refer to Patent Document 1).

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent No. 4026684 Summary of the invention

[0006] Problems to be solved by the invention

[0007] In order to accurately evaluate the growth status of plants based on image analysis, the size of each part of the plant is required. In the case where the growth status evaluation object is a fruit tree, if a blue film is placed behind it and the fruit tree is photographed with a camera from a specified distance (see Patent Document 1), an image with the size of each part can be obtained, so the growth status of the fruit tree can be accurately evaluated. However, in a place where a large number of plants are cultivated, it is actually impossible to use a camera to photograph the appearance of each plant.

[0008] The present invention has been made in view of the above-mentioned current situation, and an object of the present invention is to provide a technology capable of calculating an accurate index value related to the growth state of a plant without adding a burden to a user.

[0009] Means for solving problems

[0010] An information processing device according to one aspect of the present invention comprises: a receiving unit that receives input of a captured image, the captured image being obtained by capturing a cultivation area of ​​a plant including a rectangular region from the length direction of the rectangular region; a transformation coefficient calculation unit that determines the number of pixels in the width direction of the rectangular region in the captured image for each of a plurality of depth direction positions in the captured image input to the receiving unit, and calculates a transformation coefficient for each depth direction position by dividing the actual width of the rectangular region by each determined number of pixels; and an index value calculation unit that calculates an index value related to the growth state of the plant by analyzing the captured image using the transformation coefficient calculated by the transformation coefficient calculation unit for each depth direction position. In addition, the depth direction in the captured image refers to a direction parallel to the length direction of the rectangular region image in the captured image (the up-down direction of the captured image).

[0011] That is, the information processing device has a structure that calculates "transformation coefficients for each depth direction position" representing the width of the area represented by one pixel at each depth direction position of the photographed image based on a photographed image obtained by photographing the cultivation area of ​​the plant including the rectangular area from the length direction of the rectangular area, and analyzes the photographed image using the calculated transformation coefficients, thereby calculating an index value related to the growth state of the plant. Therefore, the index value calculated by the information processing device is more accurate than the index value obtained by analyzing the photographed image without information related to the scale according to each depth direction position. In addition, since the user of the information processing device does not need to photograph the appearance of each plant cultivated in the cultivation area, according to the information processing device, it is possible to calculate an accurate index value related to the growth state of the plant without increasing the burden on the user.

[0012] The index value calculation unit of the information processing device may also be a unit that calculates the index value by an image analysis based on artificial intelligence. The index value calculated by the index value calculation unit that performs image analysis based on artificial intelligence may also be a leaf area index (LAI: Leaf Area Index). The rectangular area included in the cultivation area of ​​the plant can be any area in which two boundary lines extending in the length direction of the area are roughly parallel. Therefore, the rectangular area may be a ridge in the cultivation area or a passage in the cultivation area. The rectangular area may also be an imaginary area defined by a plurality of columns (such as columns regularly arranged in a plastic greenhouse) arranged in the cultivation area.

[0013] In addition, an indicator value calculation method according to one aspect of the present invention prepares a captured image of the cultivation area by photographing the state of a cultivation area of ​​a plant including a rectangular area from the length direction of the rectangular area using a photographing device, determines the number of pixels in the width direction of the rectangular area in the prepared captured image for each of a plurality of depth direction positions, calculates a transformation coefficient for each depth direction position by dividing an actual width of the rectangular area by each determined number of pixels, analyzes the captured image using the transformation coefficient calculated for each depth direction position, and calculates an indicator value related to the growth state of the plant.

[0014] That is, in this index value calculation method, similarly to the above-mentioned information processing device, the captured image is analyzed using the "transformation coefficient for each depth direction position" to calculate the index value related to the growth state of the plant. Therefore, according to this index value calculation method, it is also possible to calculate an accurate index value related to the growth state of the plant without increasing the burden on the user.

[0015] Effects of the Invention

[0016] According to the present invention, it is possible to calculate an accurate index value related to the growth state of a plant without increasing the burden on the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a diagram for explaining the structure and usage of an information processing device according to one embodiment of the present invention.

[0018] Figure 2 This is a flowchart of an index value calculation process performed by a transform coefficient calculation unit and an index value calculation unit of an information processing device according to an embodiment.

[0019] Figure 3 It is a diagram for explaining the processing of steps S101 and S102 in the index value calculation processing. DETAILED DESCRIPTION

[0020] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0021] Figure 1 The structure and usage of the information processing device 10 according to one embodiment of the present invention are shown.

[0022] The information processing device 10 of the present embodiment is a device that calculates the leaf area index (LAI: Leaf Area Index) in the cultivation area of ​​each user's plant based on the captured image of the cultivation area of ​​the plant sent from the user terminal 20 of each user via the Internet 50. In addition, the user terminal 20 of the present embodiment is a smart phone installed with a program for using the information processing device 10. However, the user terminal 20 can be any device as long as it can send the captured image of the cultivation area of ​​the plant to the information processing device 10 via the Internet 50.

[0023] The information processing device 10 is a highly functional computer programmed to function as a device including a receiving unit 11, a conversion coefficient calculating unit 12, an index value calculating unit 13, and a cultivation area management database 14. Here, the receiving unit 11 corresponds to the receiving means in the present invention.

[0024] The cultivation area management database 14 is a database for managing information related to the cultivation areas of each user. The actual width W (hereinafter also referred to as passage width W) of the passage set in each cultivation area of ​​each user is set in the cultivation area management database 14. More specifically, in the cultivation area management database 14, for a user who manages one cultivation area, the passage width W of the passage set in the one cultivation area is set together with the identification information of the cultivation area. In addition, in the cultivation area management database 14, for a user who manages multiple cultivation areas, the passage width W of the passage set in each cultivation area is set together with the identification information of the cultivation area for each cultivation area.

[0025] Furthermore, authentication information (for example, user ID and password) of each user is set in the cultivation area management database 14 .

[0026] The receiving unit 11 is a functional block that receives the captured images of the plant cultivation area sent from each user terminal 20 as a processing target after user authentication. The captured images of the plant cultivation area sent from each user terminal 20 are color images captured by the user along the length direction of the passage provided in the cultivation area (see Figure 3 ). In addition, photographing the cultivation area from the length direction of the passage means photographing the cultivation area in such a way that the photographing direction (the orientation of the photographing device) is roughly consistent with the length direction of the passage. The photographed image sent to the information processing device 10 is preferably an image obtained by photographing the cultivation area from a position on the center line of the passage. However, the photographed image may also be an image obtained by photographing the cultivation area from a position slightly deviated from the center line of the passage.

[0027] In addition, when accepting the captured image as a processing target, the accepting unit 11 also obtains identification information of the cultivation area whose condition is shown in the captured image of this time from the user. Then, the accepting unit 11 uses the identification information to read the passage width W of the passage provided in the cultivation area whose condition is shown in the captured image of this time from the cultivation area management database 14. Then, the accepting unit 11 instructs the transformation coefficient calculation unit 12 to process the captured image of this time (hereinafter referred to as the captured image of the processing target) using the read passage width W (hereinafter referred to as the passage width W of the processing target).

[0028] Below, also use Figure 2 as well as Figure 3 , the functions of the transformation coefficient calculation unit 12 and the index value calculation unit 13 are explained. In addition, Figure 2 This is a flowchart of the LAI calculation process performed by the transform coefficient calculation unit 12 and the index value calculation unit 13. Figure 3 It is a diagram for explaining the processing of steps S101 and S102 in the LAI calculation processing.

[0029] Index value calculation unit 13 ( Figure 1 ) is a functional block (learned functional block) that analyzes the captured image of the processing object by image analysis based on artificial intelligence to calculate the LAI in the cultivation area whose condition is shown in the captured image of the processing object. The transformation coefficient calculation unit 12 is a functional block that calculates the transformation coefficient, which is information used by the index value calculation unit 13 when analyzing the captured image of the processing object.

[0030] The transform coefficient calculation unit 12 of this embodiment performs Figure 2 The transform coefficients are calculated by processing steps S101 to S103 (hereinafter also referred to as transform coefficient calculation processing).

[0031] That is, the transform coefficient calculation unit 12, which is instructed by the receiving unit 11 to process the processing target captured image, first searches for a path in the processing target captured image (step S101). In this step S101, the processing performed by the transform coefficient calculation unit 12 is as follows: Figure 3 As schematically shown, the processing object is captured by taking an image ( Figure 3 (A)) is divided into green parts (in Figure 3 However, the process of step S101 may also be a process of searching for a path using other algorithms (for example, retrieving a continuous area of ​​earth color from the captured image of the processing object).

[0032] The conversion coefficient calculation unit 12 that has completed the search for the path determines the number of pixels P1 to P1 in the width direction at locations of the searched path that are different in depth direction. N (Step S102). Here, the depth direction refers to the direction from bottom to top in the processing target captured image (the direction in which the distance from the camera that captured the processing target captured image becomes longer) (refer to Figure 3 (B)).

[0033] End of pixel number P1~P N The conversion coefficient calculation unit 12 determines the conversion coefficient by dividing the processing target channel width W by the number of pixels P1 to P N To find the transformation coefficients C1~C N (Step S103) As already explained (defined), the processing target passage width W is the passage width W of the passage provided in the cultivation area whose condition is shown in the processing target captured image and read from the cultivation area management database 14 by the receiving unit 11.

[0034] When the conversion coefficient calculation process (the process of steps S101 to S103) of the conversion coefficient calculation unit 12 is completed, the index value calculation unit 13 performs the following process:N The LAI is calculated by analyzing the processing target captured image and stored in the cultivation area management database 14 (step S104). The processing of step S104 performed by the index value calculation unit 13 only requires the use of the conversion coefficients C1 to C2 in the analysis of the processing target captured image. N More specifically, the processing of step S104 only needs to consider the transformation coefficients C1 to C N The processing of the captured image to be processed may be performed by analyzing the width of the area represented by one pixel at each depth direction position respectively indicated.

[0035] When the processing of step S104 is completed, the index value calculation processing for the current processing target captured image is terminated. Figure 2 of processing).

[0036] As described above, the information processing device 10 of this embodiment has a configuration for calculating conversion coefficients C1 to C2 of the width of an area represented by one pixel at each depth direction position of a captured image representing a plant cultivation area having a passageway. N , using the calculated transformation coefficients C1 to C N The captured image is analyzed to calculate the LAI. Therefore, the index value calculated by the information processing device 10 is more accurate than the index value obtained by analyzing the captured image without information about the scale of each depth direction position. In addition, since the user of the information processing device 10 does not need to take a picture of the appearance of each plant cultivated in the cultivation area, according to the information processing device 10, it is possible to calculate an accurate index value related to the growth status of the plant without increasing the burden on the user.

[0037] 《Variation Example》

[0038] The above-mentioned information processing device 10 can be modified in various ways. For example, the information processing device 10 can also be modified into a device that calculates an index value related to the growth state of the plant other than LAI. In addition, the information processing device 10 can also be modified into a device that calculates multiple index values ​​related to the growth state of the plant. The information processing device 10 can also be modified into a device that calculates an index value related to the growth state of the plant by image analysis that does not rely on artificial intelligence.

[0039] The information processing device 10 can also be transformed into a device that uses the width of the ridge in the cultivation area instead of the width of the passage in the cultivation area. In the case where there are other rectangular areas in the cultivation area (areas where two boundary lines extending in the length direction of the area are roughly parallel), the information processing device 10 can also be transformed into a device that uses the width of the cultivation area instead of the width of the passage. In addition, there are also cases where multiple columns are regularly arranged in the cultivation area. In such a case (such as the case of cultivating plants in a plastic greenhouse), the width of the imaginary rectangular area delimited by multiple columns can also be used instead of the width of the passage. In addition, in this case, the user shoots the cultivation area from the direction of the imaginary rectangular area formed by multiple columns.

[0040] Postscript

[0041] An information processing device (10), characterized by comprising:

[0042] A receiving unit (11) receives an input of a photographed image, the photographed image being obtained by photographing a cultivation area of ​​a plant including a rectangular area in a longitudinal direction of the rectangular area;

[0043] a transformation coefficient calculation unit (12) for determining the number of pixels in the width direction of the rectangular area in the captured image for each of the plurality of depth direction positions in the captured image input to the receiving unit, and calculating the transformation coefficient for each depth direction position by dividing the actual width of the rectangular area by the determined number of pixels; and

[0044] An index value calculation unit (13) calculates an index value related to the growth state of the plant by analyzing the captured image using the conversion coefficient calculated by the calculation unit for each depth direction position.

[0045] Description of symbols

[0046] 10: information processing device; 11: receiving unit; 12: conversion coefficient calculation unit; 13: index value calculation unit; 14: cultivation area management database; 20: user terminal; 50: Internet.

Claims

1. An information processing device, characterized in that: have: a receiving unit that receives an input of a captured image, the captured image being obtained by capturing a plant cultivation area including the rectangular area using the capturing device with the capturing device oriented in the same direction as the length direction of the rectangular area; a transformation coefficient calculation unit that determines the number of pixels in the width direction of the rectangular area in the captured image for each of the plurality of depth direction positions in the captured image input to the receiving unit, and calculates a transformation coefficient for each depth direction position by dividing an actual width of the rectangular area by each determined number of pixels; as well as an index value calculation unit that calculates an index value related to the growth state of the plant by analyzing the captured image using the conversion coefficient calculated by the conversion coefficient calculation unit for each depth direction position, The depth direction in the captured image is a direction parallel to the length direction of the rectangular region in the captured image. The transform coefficient for each depth-direction position of the captured image represents a width of a region represented by one pixel at each depth-direction position of the captured image.

2. The information processing device according to claim 1, characterized in that The index value calculation unit is a unit that calculates the index value through image analysis based on artificial intelligence.

3. The information processing device according to claim 2, characterized in that: The index value is the leaf area index.

4. The information processing device according to any one of claims 1 to 3, characterized in that: The rectangular area is a passage in the cultivation area.

5. The information processing device according to any one of claims 1 to 3, characterized in that: The rectangular area is a ridge within the cultivation area.

6. The information processing device according to any one of claims 1 to 3, characterized in that: The rectangular region is a virtual region defined by a plurality of pillars provided in the cultivation region.

7. A method for calculating an index value, characterized in that: By aligning the orientation of the imaging device with the length direction of the rectangular area, the imaging device is used to capture the state of the cultivation area of ​​the plant including the rectangular area, thereby preparing a captured image of the cultivation area. The number of pixels in the width direction of the rectangular area in the prepared captured image is determined for each of the plurality of depth direction positions, and a transformation coefficient for each depth direction position is calculated by dividing an actual width of the rectangular area by the determined number of pixels. By analyzing the captured image using the conversion coefficient calculated for each depth direction position, an index value related to the growth state of the plant is calculated, The depth direction in the captured image is a direction parallel to the length direction of the rectangular region in the captured image. The transform coefficient for each depth-direction position of the captured image represents a width of a region represented by one pixel at each depth-direction position of the captured image.

Images