Evaluation device and evaluation method

Abstract

To provide an evaluation device that can reduce time and effort involved in evaluation of reliability.SOLUTION: An evaluation device 10 comprises: a reception unit 12 for receiving application information for an application in relation to a carbon fixed amount of carbon fixing organisms, including a photographed image in which the carbon fixing organisms for fixing carbons to themselves are photographed, a photographed timing of the photographed image, and a position of the carbon fixing organisms photographed in the photographed image; and an evaluation unit 30 for evaluating reliability of the application information, on the basis of the photographed image, and a reference image photographed from above the carbon fixing organisms, which is an image photographed at a timing corresponding to the photographed timing of the photographed image at a position corresponding to the position.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an evaluation apparatus and an evaluation method. [Background Art]

[0002] To curb climate change, DAC (Direct Air Capture) technology, which fixes atmospheric carbon dioxide in terrestrial plants, marine algae and the like, is advancing, and a system for issuing carbon credits as compensation for the amount of carbon fixed is being established. Evaluation of the amount of carbon fixed is performed by an evaluation institution based on application documents submitted by the applicant. In this case, to prevent false applications from being mixed in, it is preferable to evaluate the reliability of the application documents and their evidence (such as photographs).

[0003] As a technology related to reliability evaluation, the abstract of Patent Document 1 describes that it "is constructed on a computer equipped with a geographic information system (GIS) engine, and includes a database 11, an acquisition unit 12 connected to the database 11, a basic data creation unit 13 connected to the database 11 and the acquisition unit 12, and a calculation unit 14 connected to the database 11." [Prior Art Documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2012-58772 [Summary of the Invention] [Problem to be Solved by the Invention]

[0005] In the technology described in Patent Document 1, an operator interprets tree species while displaying remote sensing data such as satellite photographs and on-site photographs taken at the location on the screen (paragraph 0042). The reliability of application information can be evaluated based on the operator's interpretation of tree species. However, when evaluation is performed by an operator, if there are a large number of applicants, the amount of work becomes excessive, which causes a problem that much time is required. The problem that this disclosure aims to solve is to provide an evaluation device and evaluation method that can reduce the effort involved in evaluating reliability. [Means for solving the problem]

[0006] The evaluation apparatus of this disclosure includes: a reception unit that receives application information for an application regarding the amount of carbon fixed by a carbon-fixing organism, which includes an image of a carbon-fixing organism that fixes carbon to itself, the time the image was taken, and the position of the carbon-fixing organism as seen in the image; a reference image of the carbon-fixing organism taken from above, which is an image taken at the time and position corresponding to the time the image was taken, and the position corresponding to the position; and an evaluation unit that evaluates the reliability of the application information based on the image. The evaluation unit evaluates the reliability of the application information by evaluating the consistency between the reference image and the captured image, and the consistency includes the degree of agreement between the reference image and the captured image, and the degree of agreement includes the degree of agreement of the distribution of the outlines of the carbon-fixing organisms and the degree of agreement of the colors inside the outlines for the reference image and the captured image. Other solutions will be described later in the descriptions of embodiments for carrying out the invention. [Effects of the Invention]

[0007] This disclosure provides an evaluation device and evaluation method that can reduce the effort involved in evaluating reliability. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram of the evaluation system in this disclosure. [Figure 2] This is an example of application information. [Figure 3] This is an example of a reference image. [Figure 4] This is an example of a partial image. [Figure 5] This is an example of displaying reliability on an output device. [Figure 6] This is a block diagram showing the hardware configuration of the evaluation device. [Figure 7] This is a block diagram of an evaluation system according to a different embodiment. [Figure 8] This is a block diagram of an evaluation system according to a different embodiment. [Figure 9] This is a block diagram of an evaluation system according to a different embodiment. [Figure 10]This diagram shows the relationship between the wavelength of visible light and its transmittance in water. [Figure 11] This is a block diagram of an evaluation system according to a different embodiment. [Figure 12] This is a block diagram of an evaluation system according to a different embodiment. [Figure 13] This is a block diagram of an evaluation system according to a different embodiment. [Figure 14] This is a flowchart showing the evaluation method for this disclosure. [Modes for carrying out the invention]

[0009] Hereinafter, embodiments (referred to as "models") for implementing this disclosure will be described with reference to the drawings. Within the description of one embodiment below, other embodiments applicable to that embodiment will also be described as appropriate. This disclosure is not limited to the following embodiments, and different embodiments can be combined or modified as appropriate without significantly impairing the effects of this disclosure. In addition, the same reference numerals will be used for the same components, and redundant descriptions will be omitted. Furthermore, components having the same function will be given the same name. The illustrations are for illustrative purposes only, and for illustrative purposes, the actual configuration may be changed or some components may be omitted or modified between drawings without significantly impairing the effects of this disclosure.

[0010] Figure 1 is a block diagram of the evaluation system 100 of this disclosure. The evaluation system 100 is used, for example, by evaluation organizations (e.g., national governments, local governments, corporations, etc.) that accept applications regarding carbon sequestration. However, the evaluation system 100 may also be used by organizations other than such evaluation organizations. The evaluation system 100 evaluates the confidence level 32 (Figure 5) of the application information 11 (Figure 2) submitted by the applicant, based on information such as reference images 24 (described later). The application information 11, as described in detail later, includes the time of shooting 111 (Figure 2), location 112 (Figure 2), and captured image 113 (Figure 2). The application information may further include the type of carbon-fixing organism (described later) (not shown), the amount of carbon sequestration by the carbon-fixing organism (not shown), etc.

[0011] An evaluation system 100 comprises an evaluation device 10 and an output device 34. The evaluation device 10 comprises a reception unit 12, a database 18, an extraction unit 26, and an evaluation unit 30.

[0012] The reception unit 12 is configured to receive application information 11 (FIG. 2) for an application relating to the amount of carbon fixed by a carbon-fixing organism. A carbon-fixing organism is an organism that fixes carbon derived from carbon dioxide into itself, and exists on water, in water, on land, or the like, for example. Specific examples thereof include seaweed, seagrass, mangrove forests, forests, and the like. Above all, the carbon-fixing organism is preferably an organism that fixes blue carbon and exists in or on the water of seawater, brackish water, or the like (brackish water).

[0013] Blue carbon is carbon fixed by the action of marine organisms. Carbon-fixing organisms that fix blue carbon exist at locations where water is present, such as watersides and underwater. For this reason, the position where the carbon-fixing organism exists may change due to water flow, or the topography around the carbon-fixing organism may change. Therefore, if an operator attempts to evaluate the reliability 32 of the application information 11 (FIG. 2) including the captured image 113 (FIG. 2) of the carbon-fixing organism, accurate evaluation cannot be performed because the position of the carbon-fixing organism and the surrounding topography may change. Therefore, evaluation performed by the evaluation unit 30 provided in the evaluation device 10 enables the reliability 32 to be evaluated with high accuracy based on a unified standard.

[0014] An application relating to the amount of carbon fixed is not particularly limited as long as it is an application for treating the amount of carbon fixed as one piece of information. In the example of the present disclosure, the application is an application for obtaining carbon credits based on the amount of carbon fixed by a carbon-fixing organism. As will be described in detail later, a carbon credit value, which expresses carbon credits as a monetary amount, is determined according to the magnitude of the amount of carbon fixed.

[0015] Application information 11 is received, for example, by input to the reception unit 12 by the applicant, the person in charge of the submission destination of the application information 11 (e.g., the aforementioned evaluation organization), or the like. The input is performed via an input device (not shown) such as a mouse, a keyboard, or a scanner, for example. In addition, the input may be direct input to the evaluation device 10, input from a screen (not shown) on a network (not shown) connected to the evaluation device 10, or input from a printed matter (not shown) on which the application information 11 is printed.

[0016] Figure 2 is an example of the application information 11. The application information 11 includes a captured image 113 of a carbon-fixing organism, a capturing time 111 of the captured image 113, and a position 112 of the carbon-fixing organism captured in the captured image. In Figure 2, the capturing date and time (capturing date and capturing time) are shown as an example of the capturing time 111, but the capturing time 111 may be only the capturing date, only the capturing month, only the season, or the like. In the example of Figure 2, the captured image 113 is an image obtained by capturing a coastal area where the carbon-fixing organism indicated by hatching exists in water (e.g., the sea). By including the captured image 113 that captures the carbon-fixing organism, the type and the area of the existing region of the carbon-fixing organism can be determined, and the density of carbon fixation amount that varies depending on the type of the carbon-fixing organism can be determined based on the captured image 113. The density referred to herein is the carbon fixation amount per unit area by the carbon-fixing organism in the captured image 113.

[0017] The captured image 113 is, for example, an image captured from above the water surface, but there is no limitation as long as it is an image by which the existing region of the carbon-fixing organism can be grasped, such as an image captured underwater, an image captured from the side of the carbon-fixing organism, or the like.

[0018] The position of the carbon-fixing organism includes GPS coordinates, for example. By including GPS coordinates (latitude, longitude, position in the height direction if necessary, etc.), the position of the carbon-fixing organism can be specified at any position on the Earth. However, the position is not limited to GPS coordinates. In addition, the carbon-fixing organism is usually captured with a certain extent on the captured image. Therefore, the position does not need to be a single position (e.g., one GPS coordinate) and may be specified by a plurality of positions (e.g., a plurality of GPS coordinates).

[0019] Returning to Figure 1, database 18 stores reference images 24 (Figure 3) of carbon-fixing organisms taken from above, along with the time of acquisition. Reference images 24 are images of carbon-fixing organisms, for example, that exist in water, taken from above (preferably above the water surface in the case of aquatic carbon-fixing organisms). Such images can be images taken using unmanned aerial vehicles (drones, etc.) or manned aerial vehicles (aircraft, etc.) (e.g., aerial photographs), satellite images, etc.

[0020] Figure 3 shows an example of a reference image 24. The reference image 24 includes multiple images 241 of the same location, taken at different times. The shooting times can be set, for example, at predetermined intervals of days, months, or seasonally. In the example in Figure 3, the reference image 24 is a collection of multiple satellite images taken of the same location every two months. The reference image 24 may be a collection of multiple independent images for each shooting location, or it may be prepared as a single image by integrating all the captured images.

[0021] Reference image 24 is an image that includes a first part 242 which captures the location of carbon-fixing organisms, and a second part 243 which captures a second location other than the location corresponding to the first part 242. In the example in Figure 3, the second part 243 includes land 244 and sea 245. The carbon-fixing organisms shown in the first part 242 are located in sea 245.

[0022] Returning to Figure 1, the extraction unit 26 extracts a partial image 25 (Figure 4) containing the first part 242 from the reference image 24 (Figure 3) as a new reference image 24. The extraction unit 26 can exclude at least a portion of the second part 243 other than the first part 242 that corresponds to carbon-fixing organisms. This simplifies the calculations for comparison when the evaluation unit 30 (described later) performs the comparison.

[0023] Figure 4 shows an example of a partial image 25. In this disclosure, a partial image 25 obtained by extracting a part of a reference image 24 is also considered a reference image 24. Extraction is performed by extracting one image 241 (Figure 3) from a collection of multiple images 241, preferably in the vicinity of, and more preferably matching, the shooting time 111 (Figure 2) of the captured image 113 (Figure 2). Furthermore, image processing related to extraction can also be performed on the extracted single image 241 so as to include a first part 242 and remove at least a part of the second part 243. Such image processing is preferably performed in such a way that the amount of the second part 243 that is captured is reduced. This makes it possible to increase the proportion of the first part 242 containing carbon-fixing organisms in the partial image 25, which also serves as the reference image 24, and improve the recognition accuracy of carbon-fixing organisms, as will be described in detail later.

[0024] Returning to Figure 1, the evaluation unit 30 evaluates the reliability 32 (Figure 5) of the application information 11 (Figure 2) based on the reference image 24 (Figure 5), which was taken at the time corresponding to the time 111 (Figure 2) of the captured image 113 (Figure 2), and at the location corresponding to the location 112 (Figure 2), and the captured image 113. In this disclosure, the evaluation unit 30 evaluates the reliability 32 of the application information 11 by evaluating the consistency between the reference image 24 and the captured image 113. The higher the consistency, the greater the degree of agreement between the reference image 24 and the captured image 113, and the more reliable the application information is considered to be. Therefore, the evaluation unit 30 can evaluate the reliability 32 of the application information by performing a consistency evaluation in this manner.

[0025] If the consistency between the reference image 24 and the captured image 113 is high, it is considered highly likely that the applicant has submitted an application with correct information, and the confidence level of the application information 11 is considered high (32). On the other hand, if the consistency between the reference image 24 and the captured image 113 is low, it is considered likely that the applicant has submitted an application with false information, and the confidence level of the application information 11 is considered low (32). Both consistency and confidence level 32 can be expressed numerically. For example, if consistency is expressed as a numerical value from 0 to 100%, the numerical value of consistency can be used as the numerical value of confidence level 32 from 0 to 100%. For example, by setting the application information 11 with the highest consistency to 100% and the application information 11 with the lowest reliability to 0%, the confidence level 32 can be expressed as a relative numerical value. The numerical value of confidence level 32 can be used, for example, for scoring applicants.

[0026] Consistency includes, for example, the degree of agreement between the reference image 24 and the captured image 113. The higher the degree of agreement between these images, the more consistent they are, that is, the more consistent the application information 11 is with the information held by the evaluation system 100. Therefore, the confidence level 32 can be evaluated by including the degree of agreement.

[0027] In the examples of this disclosure, the degree of agreement includes the degree of agreement of the distribution of carbon-fixing organism contours and the degree of agreement of the colors within the contours for reference image 24 and captured image 113. By using the distribution and color agreement, consistency can be evaluated even if at least one of the color or shape (area of ​​existence) of the carbon-fixing organisms changes over time, as will be described in detail later. The distribution is the contour of the carbon-fixing organisms in reference image 24 and captured image 113, and therefore indicates the area of ​​existence of the carbon-fixing organisms. Accordingly, the degree of agreement of the distribution is determined based on the contours of the carbon-fixing organisms in the first part 242 (Figure 4) which includes the carbon-fixing organisms. Thus, in the examples of this disclosure, for example, consistency is evaluated based on the degree of agreement of the distribution, i.e., the degree of agreement of the distribution, in reference image 24 and captured image 113, to what extent the distribution on reference image 24 including the carbon-fixing organisms matches.

[0028] The degree of agreement in the distribution can be determined by any method. For example, as shown in Figures 2 and 5, the positions of the first portion 242 containing carbon-fixing organisms often do not perfectly match in the reference image 24 and the captured image 113. Therefore, the evaluation unit 30 determines the location where the superposition matches best by performing image processing such as fine-tuning, rotation, and scaling of one or both of the reference image 24 or the captured image 113 in the x and y directions. The evaluation unit 30 then determines the degree of agreement at the determined location. For example, within the range of position 112 (Figure 2), the absolute difference in brightness for each pixel of the reference image 24 and the captured image 113 is calculated and accumulated. If the accumulated value is close to 0, the degree of agreement can be evaluated as maximum. Conversely, if the accumulated value is large, the degree of agreement can be judged as low.

[0029] Furthermore, the evaluation unit 30 determines the degree of color matching based, for example, on the color information of the reference image 24 and the captured image 113. That is, since the shooting times of the reference image 24 and the captured image 113 are the same or close, even carbon-fixing organisms whose colors change depending on the season, time of year, etc., will have the same hue. Therefore, the degree of matching can be determined using the color information as a clue. If the hue is significantly different, the degree of matching can be judged as low, and in this case, the reliability 32 of the application information 11 (Figure 2) can be evaluated as low.

[0030] The method for determining the degree of color information matching is arbitrary. For example, the evaluation unit 30 extracts numerical information for each of the three RGB colors from the same position in both the reference image 24 and the captured image 113, and determines the degree of color information matching for this point as well. For example, within the range of position 112 (Figure 2), the absolute value of the difference for each of the three RGB colors of the numerical information of each pixel in both images is calculated and accumulated, and if the accumulated value is close to 0, the degree of matching can be evaluated as maximum. Conversely, if the accumulated value is large, it can be judged that the degree of matching is low.

[0031] Furthermore, if the captured image 113 is an image taken using an unmanned or manned aircraft (e.g., an aerial photograph), the area where carbon-fixing organisms exist can be determined, and consistency can be evaluated using the shape of the area as a clue. Similarly, in the case of high-resolution satellite images, the shape of the area can also be used as a clue.

[0032] The output device 34 outputs information regarding the confidence level 32 evaluated by the evaluation unit 30. The information regarding the confidence level 32 may be the confidence level 32 itself, the applicant who submitted the application information 11 (Figure 2) having the determined confidence level 32, or information derived from the confidence level 32. The output device 34 may be, for example, a display that shows the confidence level 32, a personal computer that compiles the confidence level 32 in a format usable by the user, or a server with a cloud that stores the confidence level 32. This allows the user to utilize the confidence level 32. The confidence level 32 may be used as is (e.g., displayed), compiled into a list for each user using, for example, spreadsheet software, or only used (e.g., displayed) by applicants with a confidence level 32 of a certain level or higher.

[0033] Furthermore, the output device 34 may output the consistency determined by the evaluation unit 30. Also, since the evaluation by the evaluation unit 30 is not necessarily perfectly accurate, the evaluation device 10 may be equipped with a correction unit (not shown) that allows the user to check and correct the contents as appropriate.

[0034] Figure 5 shows an example of the display of the confidence score 32 on the output device 34. In the example of this disclosure, the confidence score 32 is displayed as a two-dimensional plot with the degree of agreement of the distribution (contour distribution) on the horizontal axis and the degree of agreement of the color information on the vertical axis. This makes it easy to grasp the confidence score 32. For example, suppose the applicant has correctly identified the contour distribution of carbon-fixing organisms, but has applied for a type of carbon-fixing organism that can fix more carbon than the actual carbon-fixing organisms in order to apply for a higher carbon credit amount. In this case, the degree of agreement of the distribution will be high, but the degree of agreement of the color information will be low. Therefore, the confidence score 32 can be judged to be low. Furthermore, suppose the applicant has correctly identified the type of carbon-fixing organism, but has applied for a wider area of ​​existence than the actual carbon-fixing organisms in order to apply for a higher carbon credit amount. In this case, the degree of agreement of the color information will be high, but the degree of agreement of the distribution will be low. Therefore, the confidence score 32 can be judged to be low.

[0035] Furthermore, the graph in Figure 5 plots the degree of agreement in the distribution and the degree of agreement in the color information for each of the applicants A, B, C, and D's application information 11 (Figure 2). In addition, the average values ​​Av for both the degree of agreement in the distribution and the degree of agreement in the color information are also plotted.

[0036] The degree of agreement in the distribution and the degree of agreement in the color information for applicants A, B, and C are all greater than the average value Av. Therefore, the confidence level of the application information for applicants A, B, and C is high (32). On the other hand, the degree of agreement in the distribution and the degree of agreement in the color information for applicant D are smaller than the average value Av. Therefore, the confidence level of the application information for applicant D is low (32).

[0037] Using a threshold such as the average value Av as a basis for judgment, the following advantages can be obtained. For example, applicant D's application information 11 (Figure 2), which is lower than the threshold, may contain false information. Therefore, for applicant D's application, it is possible to conduct field investigations by actually going to the site or to conduct interviews with applicant D. On the other hand, the application information 11 of applicants A, B, and C, which is higher than the threshold, is less likely to contain false information. Therefore, for applicants A, B, and C's applications, field investigations etc. can be omitted, and the decision on whether or not to approve the application can be made based solely on desk judgment. In this way, by determining which applications should be prioritized for investigation based on the results obtained by the evaluation device 100, limited investigation capabilities can be used effectively.

[0038] If we were to quantify a confidence score of 32 based on the degree of agreement between the distribution and the degree of agreement between the color information, for example, it could be quantified by multiplying the degree of agreement between the distribution and the degree of agreement between the color information.

[0039] As described above, according to the embodiments shown in Figures 1 to 5, the reliability 32 of the application information 11 can be efficiently evaluated upon receipt of the application information 11. This reduces the effort required to evaluate the reliability 32. Furthermore, by using a reference image 24 (e.g., satellite image), which is a different information source from the application information 11, even if there are multiple applicants, they can be evaluated fairly according to the same criteria.

[0040] Figure 6 is a block diagram showing the hardware configuration of the evaluation device 10. The evaluation device 10 may also be implemented in software by having a processor such as a CPU 252 interpret and execute programs that realize each function. The evaluation device 10 includes, for example, a memory 251, a CPU 252, a storage device 253 (SSD, HDD, etc.), a communication device 254, and an I / F 255. Information such as programs, tables, and files that realize each function can be stored in memory, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC (Integrated Circuit) card, an SD (Secure Digital) card, or a DVD (Digital Versatile Disc), in addition to being stored in the HDD. Furthermore, information can be stored in the cloud, or calculations and other processing can be performed in the cloud.

[0041] Figure 7 is a block diagram of an evaluation system 100 according to another embodiment. The evaluation device 10 shown in Figure 7 includes a selection unit 42 in addition to the configuration of the evaluation system 100 shown in Figure 1.

[0042] The selection unit 42 selects multiple reference images 24 from the database 18 at the same location 112 (Figure 2) but at different capture times 111 (Figure 2), and then selects a reference image 24 from the selected reference images 24 in which carbon-fixing organisms can be recognized. By including the selection unit 42, reference images 24 in which obstacles such as clouds are above the carbon-fixing organisms and the carbon-fixing organisms cannot be recognized can be excluded. As a result, the confidence level 32 can be evaluated using reference images 24 in which carbon-fixing organisms can be recognized. In the evaluation device 10 shown in Figure 7, the selection unit 42 performs selection on the reference images 24 extracted by the extraction unit 26.

[0043] As described above, the database 18 stores past reference images 24 (Figure 3). Since the reference images 24 are images taken from above carbon-fixing organisms, if there are clouds between the ground and the sea surface, it may not be possible to precisely capture the ground and sea surface in the desired range, or they may not be captured at all. This is particularly noticeable when the reference images 24 are satellite images. In such cases, the evaluation accuracy of the evaluation unit 30 decreases. Furthermore, if the carbon-fixing organisms are, for example, marine algae, if there are whitecaps on the sea surface, or if sunlight is reflected off the water surface, it may not be possible to precisely capture, or they may not be captured at all, the algae in the sea and on the seabed in the desired range. In such cases as well, the evaluation accuracy of the evaluation unit 30 decreases.

[0044] Therefore, in the embodiment shown in Figure 7, a selection unit 42 is used to suppress such a decline in evaluation accuracy. It is preferable that the selection by the selection unit 42 is performed automatically. As described above, the selection unit 42 selects a plurality of reference images 24 from different time periods. At this time, the selection unit 42 selects a reference image 24 that was photographed at the same location as the location 112 (Figure 2) of the application information 11. It is preferable that the selection unit 42 selects a reference image 24 that was photographed at a time as close as possible (preferably matching) the photographing time 111 (Figure 2) of the application information 11.

[0045] The selection unit 42 selects a reference image 24 in which carbon-fixing organisms can be recognized by excluding images that show temporary obstacles above the carbon-fixing organisms. This allows for the detection of temporary obstacles by comparing multiple reference images 24, and the exclusion of reference images 24 that show such obstacles. Obstacles include, for example, clouds, whitecaps, and reflections on the water surface. In the example of this disclosure, the selection unit 42 selects a reference image 24 in which no obstacles are present, or in which the obstacles are few enough to recognize the carbon-fixing organisms.

[0046] Clouds, whitecaps, and water surface reflections are temporary phenomena, and in the examples of this disclosure, the characteristics of all being high in brightness and white are utilized. Specifically, the selection unit 42 can determine that a pixel at the same location is cloud, whitecaps, or water surface reflection if its brightness is temporarily high among several different time periods. However, the method for determining the presence or absence of an obstacle is not limited to this example.

[0047] Thus, according to the embodiment shown in Figure 7, the effects of temporary occurrences caused by natural phenomena such as clouds, whitecaps, and water surface reflections that can realistically occur can be reduced, thereby improving the evaluation accuracy of the evaluation unit 30.

[0048] Figure 8 is a block diagram of an evaluation system 100 according to another embodiment. The evaluation device 10 shown in Figure 8 includes an estimation unit 48 in addition to the configuration of the evaluation system 100 shown in Figure 7. In the example shown in Figure 8, the application information 11 (Figure 2) further includes the type of carbon-fixing organism (e.g., seaweed, seagrass, mangrove forest, forest, etc.).

[0049] The estimation unit 48 selects multiple reference images 24 from the database 18 at the same location 112 (Figure 2) but at different shooting times 111 (Figure 2). Furthermore, the estimation unit 48 estimates the types of carbon-fixing organisms captured in the selected reference images 24 based on the color information of each carbon-fixing organism captured in the reference images 24. By including the estimation unit 48, the types of carbon-fixing organisms captured in the reference images 24 can be estimated and used to evaluate the confidence level 32 by comparing them with the types included in the application information 11 (Figure 2). In the evaluation device 10 shown in Figure 8, the estimation unit 48 performs estimation on the reference images 24 selected by the selection unit 42.

[0050] As described above, the evaluation unit 30 evaluates the consistency between the reference image 24 (Figure 4) and the captured image 113 (Figure 2). At the same time, the evaluation unit 30 also evaluates the consistency between the types of carbon-fixing organisms in the reference image 24, as inferred by the estimation unit 48, and the types of carbon-fixing organisms captured in the captured image 113 (Figure 2) included in the application information 11. Therefore, the evaluation unit 30 further incorporates the consistency of the types into the evaluation and determines the confidence level 32.

[0051] Each species of carbon-fixing organism has a unique color, and some carbon-fixing organisms, such as red algae, change their surface color depending on the season. By using these unique colors and seasonal color changes as RGB numerical information of the image, the estimation unit 48 can estimate the type of carbon-fixing organism.

[0052] Thus, according to the embodiment shown in Figure 8, the type of carbon-fixing organism can be determined, and it is possible to ascertain whether the type of carbon-fixing organism applied for by the applicant is appropriate.

[0053] Figure 9 is a block diagram of an evaluation system 100 according to another embodiment. The evaluation device 10 shown in Figure 9 includes a correction unit 58 in addition to the configuration of the evaluation system 100 shown in Figure 8.

[0054] The correction unit 58 generates a new reference image 24 from the reference image 24, correcting for the absorption characteristics due to water. By including the correction unit 58, the colors of carbon-fixing organisms that are not properly represented on the reference image 24 due to the absorption characteristics of water can be accurately identified. In the evaluation device 10 shown in Figure 9, the correction unit 58 corrects the reference image 24 extracted by the extraction unit 26.

[0055] Figure 10 shows the relationship between the wavelength of visible light and its transmittance in water. The solid line graph shows the relationship between wavelength and transmittance in water when the optical path length is 10 m (i.e., when the light source is located at a depth of 5 m). Note that the shape of the solid line graph may change if the optical path length changes. Figure 10 also illustrates the wavelength ranges of red (R), green (G), and blue (B).

[0056] As shown in Figure 10, the transmittance of visible light through water (vertical axis) changes depending on the wavelength of visible light (horizontal axis). Therefore, if the carbon-fixing organism is, for example, algae present in the sea, a color different from its original color will be captured as the reference image 24. Furthermore, the color of the carbon-fixing organism in the reference image 24 differs depending on the water depth at which the carbon-fixing organism is located. This is because the presence of a water layer causes differences in the absorption rate of light depending on the wavelength. This difference may affect the species determination by the estimation unit 48 (Figure 9).

[0057] Therefore, the correction unit 58 (Figure 9) corrects the RGB color information, taking into account the difference in transmittance. In the graph shown in Figure 10, as described above, the optical path length is 10m, which corresponds to a water depth of about 5m when considering the round trip journey of sunlight. This water depth is close to the depth where seaweed is abundant. Here, the transmittance of red (R) is about 10% of that of blue (B). Accordingly, the correction unit 58 applies a process to the RGB color information by multiplying it by the reciprocal of this underwater transmittance to determine RGB color information that is close to the original color.

[0058] If the image 113 (Figure 2) in application information 11 (Figure 2) is an image taken from above a carbon-fixing organism, then if the water depth is the same, the absorption of light due to the presence of water will be the same and there will be no significant effect. However, if the image 113 is an underwater photograph (where light absorption in the water layer occurs only in one direction), or if it is a photograph taken on land where there is no water layer (no light absorption in the water layer), the processing by the correction unit 58 can improve the evaluation accuracy of the evaluation unit 30.

[0059] The water depth at which carbon-fixing organisms exist varies depending on the type of organism. Therefore, if the application information includes the type of carbon-fixing organism, the water depth should be determined according to the applied-for type. Based on the determined water depth, the relationship shown in Figure 10 can be determined, and the correction unit 58 can correct the color information. In this case, it is preferable that the evaluation device 10 includes a database that associates the type of carbon-fixing organism with the water depth.

[0060] Furthermore, in the case of the ocean, the water depth (distance from the water surface) of carbon-fixing organisms changes with the ebb and flow of the tides. Therefore, for example, the water depth may be determined by considering the tidal level at the time of filming 11 (Figure 2).

[0061] Thus, according to the embodiments shown in Figures 9 and 10, the accuracy of predicting the types of carbon-fixing organisms can be improved.

[0062] Figure 11 is a block diagram of an evaluation system 100 according to another embodiment. The evaluation device 10 shown in Figure 11 includes, in addition to the configuration of the evaluation system 100 shown in Figure 9, a roughness determination unit 68 and a fixation amount determination unit 72. Also, in the example shown in Figure 11, the application information 11 (Figure 2) includes the amount of carbon fixed by carbon-fixing organisms.

[0063] The roughness determination unit 68 determines the roughness based on the reference image 24. Roughness is an index that indicates the density of carbon-fixing organisms present in the region where carbon-fixing organisms exist in the reference image 24. In the example of this disclosure, roughness is a relative value where 1 represents a state where the seabed is not visible due to the density, and 0 represents a state where the seabed can be seen because no carbon-fixing organisms are present. In the evaluation device 10 shown in Figure 11, the roughness is determined by the roughness determination unit 68 for the reference image 24 selected by the selection unit 42.

[0064] Roughness is determined for each unit area (for example, several meters by several meters in actual environments) within the area where carbon-fixing organisms exist. Therefore, in actual environments, for example, in vast areas of several tens of meters by several tens of meters, there are multiple roughness values ​​determined for each unit area in the reference image 24.

[0065] In the example of this disclosure, the roughness determination unit 68 calculates roughness based on a reference image 24 and the types of carbon-fixing organisms visible in the reference image 24. Once the types are determined, the RGB color information of the determined type of carbon-fixing organism is known. Areas in the reference image 24 where RGB colors are visible are considered to have a high density of carbon-fixing organisms, and therefore the roughness is high, close to 1. For example, a roughness of 1 means that the seabed is not visible at all in the reference image 24. On the other hand, areas in the reference image 24 where RGB colors are not visible are considered to be places where no carbon-fixing organisms exist, i.e., the ground in the case of land, or the seabed in the case of the ocean. Therefore, the roughness is low, close to 0. If the RGB colors visible in the reference image 24 are in between these two ranges, it can be determined that there are some carbon-fixing organisms present, but they are not dense, and the roughness value will be between 0 and 1.

[0066] The fixation amount determination unit 72 determines the amount of carbon fixed by carbon-fixing organisms in the region of existence in the reference image 24, based on the roughness determined by the roughness determination unit 68. By providing the above-mentioned roughness determination unit 68 and fixation amount determination unit 72, the amount of carbon fixed by carbon-fixing organisms can be determined based on the reference image 24. As a result, the evaluation unit 30 can evaluate the consistency between the amount of carbon fixed based on the reference image 24 and the amount of carbon fixed included in the application information. As a result, the reliability score 32 can be evaluated taking into account the evaluation results, and the accuracy of the reliability score 32 can be improved. In the evaluation device 10 shown in Figure 11, the fixation amount determination unit 72 receives information on the type of carbon-fixing organism estimated by the estimation unit 48, the roughness determined by the roughness determination unit 68, and the amount of carbon fixed included in the application information 11 (Figure 2).

[0067] The amount of carbon fixed increases with the size of the region in the image where carbon-fixing organisms exist (the region of existence). Here, the region of existence is defined as a closed region obtained by dividing the image into closed areas using the outermost carbon-fixing organisms. Furthermore, the amount of carbon fixed increases with the density of carbon-fixing organisms within the region of existence in the image. That is, in the region of existence, the denser the organisms are (the less visible the seabed), the higher the density; conversely, the sparser the organisms are (the more visible the seabed), the lower the density. Also, the amount of carbon fixed varies depending on the type of carbon-fixing organism. Therefore, the amount of carbon fixed can be determined by determining the roughness of the region.

[0068] In the example of this disclosure, the fixation amount determination unit 72 determines a fixation coefficient that indicates the extent to which carbon is fixed, based on the location and type of carbon-fixing organisms. The fixation amount determination unit 72 also determines the area of ​​the region where the carbon-fixing organisms are located, based on their location. As described above, roughness usually differs from unit to unit. That is, in a region where multiple unit regions are composed, the roughness differs for each unit region and therefore has a distribution. For this reason, the fixation amount determination unit 72 can determine the amount of carbon fixed by multiplying the area of ​​the region where the carbon fixed is located, the roughness, and the fixation coefficient. The amount of carbon fixed is provided to the output device 34. It is desirable that the output device 34 also simultaneously display the amount of carbon fixed requested by the applicant from the application information.

[0069] Thus, according to the embodiment shown in Figure 11, the amount of carbon fixation can be efficiently determined based on the reference image 24. Furthermore, by taking into account the requested amount of carbon fixation, such as the amount of carbon fixation requested by the applicant being output (e.g., displayed) on the output device 34, a comprehensive judgment can be made together with the reliability 32.

[0070] Figure 12 is a block diagram of an evaluation system 100 according to another embodiment. The evaluation device 10 shown in Figure 12 includes, in addition to the configuration of the evaluation system 100 shown in Figure 11, a selection unit 76, a bottom color determination unit 80, and a distribution determination unit 84. Also, in the example shown in Figure 12, the application information 11 (Figure 2) includes the types of carbon-fixing organisms.

[0071] The selection unit 76 selects a reference image 24 from the database 18 that was taken at a time when the color of the seabed below the carbon-fixing organisms of the same type as those captured in the image 113 (Figure 2) could be distinguished. The seabed color determination unit 80 determines the color of the seabed based on the reference image 24 selected by the selection unit 76. The color of the seabed is determined based on the application information 11 (Figure 2), which will be described in detail later.

[0072] The distribution determination unit 84 determines the roughness for each unit region constituting the presence area, based on the color of the carbon-fixing organisms in the reference image 24 and the color of the seabed determined by the seabed color determination unit 80, and also determines the distribution of roughness for each unit region within the presence area. The distribution of roughness is input to the roughness determination unit 68.

[0073] By comprising a selection unit 76, a bottom color determination unit 80, and a distribution determination unit 84, the distribution of roughness in the presence area can be determined. This allows for the detection of false application information 11 (Figure 2) that states the roughness of the entire presence area is 1, even though the roughness at the edges of the presence area is actually less than 1. Furthermore, the amount of carbon fixed differs depending on whether the roughness is 1 in all unit areas constituting the presence area or 0.5 in all unit areas. The amount of unit fixation can be determined according to the roughness of each unit area. This allows for the evaluation of the consistency between the amount of carbon fixed requested by the applicant and the amount of carbon fixed based on the reference image 24, enabling a comprehensive judgment that takes the amount of carbon fixed into account, along with the confidence level 32.

[0074] In the evaluation system 100 shown in Figure 11 above, if the RGB colors observed in the reference image 24 are between these values, it can be determined that carbon-fixing organisms are present to some extent but not densely, and the roughness value will be between 0 and 1. In the evaluation system 100 shown in Figure 12, roughness is determined more accurately by using information about the color of the seabed. The RGB values ​​in the unit area constituting the presence area are the interpolation (intermediate value) between the RGB values ​​of the color corresponding to the type of carbon-fixing organism and the RGB values ​​given as the color of the seabed. Therefore, the distribution determination unit 84 determines the roughness by calculating the ratio of how close it is to the color of the carbon-fixing organism or the seabed.

[0075] The roughness (density in the area of ​​presence) of carbon-fixing organisms can change depending on the season or other time of year. That is, the amount of carbon-fixing organisms present in the area of ​​presence in the reference image 24 increases or decreases depending on the time of year. Therefore, at one time, there may be a very large number of carbon-fixing organisms in the area of ​​presence, but at another time, there may be almost no carbon-fixing organisms in the area of ​​presence. Therefore, the selection unit 76 preferably selects a reference image 24 taken at a time when the color of the seabed can be discerned, such that there are almost no carbon-fixing organisms. The reference image 24 taken at a time when the color of the seabed can be discerned can be determined based on the types of carbon-fixing organisms included in the application information 11 (Figure 2). However, the selection unit 76 may also detect changes in the roughness of carbon-fixing organisms by comparing multiple reference images 24 at the same location, for example. This allows the selection unit 76 to select a reference image 24 taken at a time when the color of the seabed can be discerned.

[0076] As described above, the seabed color determination unit 80 determines the color of the seabed based on the reference image 24 selected by the selection unit 76. The seabed color can be determined as the portion of the selected reference image 24 that is not the color corresponding to the type of carbon-fixing organism included in the application information 11 (Figure 2). However, the seabed color may also be determined by the color of the portion that does not change when multiple reference images 24 are compared by the selection unit 76. In addition, the seabed color may be determined by taking into account, for example, the water depth to the seabed. To ensure consistent conditions, it is preferable to determine the seabed color based on a reference image 24 from a time when carbon-fixing organisms are not present at the same location as the area where carbon-fixing organisms exist, but the seabed color may also be determined based on a reference image 24 from which carbon-fixing organisms are not present. This makes the determination easy.

[0077] Thus, according to the embodiment shown in Figure 12, the amount of carbon fixation can be efficiently determined more accurately based on the reference image 24.

[0078] Figure 13 is a block diagram of an evaluation system 100 according to another embodiment. The evaluation device 10 shown in Figure 13 includes a carbon credit determination unit 88 in addition to the configuration of the evaluation system 100 shown in Figure 12. The carbon fixation amount determination unit 72 determines the amount of carbon fixed by carbon-fixing organisms in the region where carbon-fixing organisms are present in the reference image 24. The amount of carbon fixed determined by the carbon fixation amount determination unit 72 is input to the carbon credit determination unit 88. The application information 11 (Figure 2) includes the amount of carbon credits requested by the applicant.

[0079] The carbon credit determination unit 88 determines the carbon credit amount based on the determined carbon sequestration amount. By including the carbon credit determination unit 88, the carbon credit amount can be determined without requiring any action from the user. This allows for comparison with the carbon credit amount requested by the applicant, and the evaluation unit 30 can evaluate the reliability 32 taking the carbon credit amount into account. The determined carbon credit amount is output (for example, displayed) to the output device 34.

[0080] The carbon credit determination unit 88 further determines the amount obtained by subtracting an amount corresponding to the determined amount of carbon sequestration from the determined carbon credit amount. This allows the applicant to be paid the amount after deducting the amount corresponding to the amount of carbon sequestration. The amount corresponding to the amount of carbon sequestration can be collected, for example, as a usage fee for the evaluation system 100. Also, for example, the amount to be deducted can be set higher for larger amounts of carbon sequestration and lower for smaller amounts of carbon sequestration. Applicants who seize a lot of carbon are thought to have large-scale facilities and businesses and therefore large transaction volumes. On the other hand, applicants who seize a little carbon are thought to have small-scale facilities and businesses and therefore small transaction volumes. Therefore, by setting the amount to be deducted to be higher for larger amounts of carbon sequestration and lower for smaller amounts of carbon sequestration, it can be increased or decreased according to the transaction volume, contributing to fairness among applicants.

[0081] Figure 14 is a flowchart illustrating the evaluation method of the present disclosure. The evaluation method of the present disclosure can be performed, for example, by the evaluation apparatus 10 of each embodiment described above. Therefore, the matters described for the evaluation apparatus 10 of each embodiment described above can be directly applied to the evaluation method shown in Figure 14.

[0082] The evaluation method of this disclosure includes an acceptance step S1 and an evaluation step S2. Acceptance step S1 is a step of receiving application information 11 (Figure 2) for an application regarding the amount of carbon fixed by carbon-fixing organisms. The application information 11 includes, as described above, a photographed image 113 (Figure 2) of a carbon-fixing organism that fixes carbon to itself, the time 111 (Figure 2) when the photographed image 113 was taken, and the position 112 (Figure 2) of the carbon-fixing organism shown in the photographed image 113. Acceptance step S1 can be performed by the acceptance unit 12.

[0083] Evaluation step S2 is a step in which the reliability 32 (Figure 2) of the application information 11 is evaluated based on the captured image 113 and the reference image 24, which is an image taken at the time 111 corresponding to the time and location 112 corresponding to the time 112 of the captured image 113. The reference image 24 is an image of a carbon-fixing organism taken from above, as described above. Evaluation step S2 can be performed by the evaluation unit 30.

[0084] According to the evaluation method disclosed herein, the reliability 32 of the application information 11 can be efficiently evaluated in evaluation step S2 upon receipt of the application information 11 in reception step S1. This reduces the effort required for evaluating the reliability 32. [Explanation of Symbols]

[0085] 10 Evaluation device 100 Evaluation System 11 Application information 111 Filming period 112 positions 113 Photographed images 12 Reception Department 18 Databases 24 Reference Sonar 241 images 242 Part 1 243 Part 2 244 Land 245 Sea 25 partial images 26 Extraction part 30 Evaluation Department 32. Confidence level 34 Output device 42 Selection Section 48 Guessing part 58 Correction section 68 Roughness determination section 72 Fixed amount determination section 76 Selection Section 80 Bottom color determination section 84 Distribution determining part 88 Carbon Credit Determination Section S1 Registration Step S2 Evaluation Step

Claims

1. A reception unit that receives application information for an application regarding the amount of carbon fixed by carbon-fixing organisms, including a photograph of a carbon-fixing organism that fixes carbon to itself, the date the photograph was taken, and the position of the carbon-fixing organism in the photograph. The system comprises a reference image of the carbon-fixing organism taken from above, which is an image taken at a time and position corresponding to the time the aforementioned image was taken, and an evaluation unit that evaluates the reliability of the application information based on the aforementioned image, The evaluation unit evaluates the reliability of the application information by evaluating the consistency between the reference image and the captured image. The aforementioned consistency includes the degree of agreement between the reference image and the captured image, The degree of agreement includes the degree of agreement of the distribution of the outlines of the carbon-fixing organisms and the degree of agreement of the colors within the outlines for the reference image and the captured image. An evaluation device characterized by the following features.

2. The aforementioned reference image is an image that includes a first portion which captures the location of the carbon-fixing organism and a second portion which captures a second location other than the location corresponding to the first portion. The system includes an extraction unit that extracts a partial image, including the first portion, from the aforementioned reference image as a new reference image. The evaluation apparatus according to feature 1.

3. A database storing the aforementioned reference images along with the time they were taken, The system includes a selection unit that selects from the database an image in which the carbon-fixing organism can be recognized from among multiple images taken at the same location as the carbon-fixing organism in the captured image but at different times. The evaluation unit uses the image selected by the selection unit as the reference image to evaluate the reliability. The evaluation apparatus according to feature 1.

4. The selection unit selects images in which the carbon-fixing organism can be recognized by excluding images that show temporary obstacles above the carbon-fixing organism. The evaluation apparatus according to feature 3.

5. The aforementioned application information includes the type of carbon-fixing organism, A database storing the aforementioned reference images along with the time they were taken, The system includes an estimation unit that selects from the database a plurality of reference images taken at the same location as the carbon-fixing organism captured in the captured image but at different times, and estimates the type of carbon-fixing organism captured in the reference image based on the color information of the carbon-fixing organism captured in each of the selected plurality of reference images, The evaluation unit compares the type of carbon-fixing organism estimated by the estimation unit with the type of carbon-fixing organism included in the application information to evaluate whether the type of carbon-fixing organism included in the application information is appropriate, and evaluates the reliability taking this evaluation into account. The evaluation apparatus according to feature 1.

6. The aforementioned location includes GPS coordinates. The evaluation apparatus according to feature 1.

7. The aforementioned application information includes the type of carbon-fixing organism, The correction unit determines the water depth at which the carbon-fixing organisms of the type included in the application information exist from the reference image, and generates an image with color information corrected using the visible light transmittance corresponding to the determined water depth. The evaluation unit uses the image generated by the correction unit as the reference image to evaluate the reliability. The evaluation apparatus according to feature 1.

8. The aforementioned application information includes the amount of carbon fixed by the carbon-fixing organisms, A roughness determination unit determines roughness, which is an indicator of the density of carbon-fixing organisms present in the region where the carbon-fixing organisms are located, by determining the color information possessed by the carbon-fixing organisms from the types of carbon-fixing organisms shown in the reference image and detecting whether or not the determined color information is present in the reference image. The system includes a fixation amount determination unit that determines the amount of carbon fixed by the carbon-fixing organism in the region of existence in the reference image based on the determined roughness, The evaluation unit evaluates whether the amount of carbon fixed determined by the carbon fixed amount determination unit matches the amount of carbon fixed included in the application information, and evaluates the reliability taking this evaluation into account. The evaluation apparatus according to feature 1.

9. The aforementioned application information includes the type of carbon-fixing organism and the amount of carbon fixed by the carbon-fixing organism. A database storing the aforementioned reference images along with the time they were taken, A selection unit selects from the database a reference image taken at a time when it is possible to distinguish the color of the seabed below the carbon-fixing organism of the same type as the carbon-fixing organism captured in the aforementioned image. A bottom color determination unit determines the color of the bottom of the sea based on the reference image selected by the selection unit, The aforementioned reference image includes a distribution determination unit that determines, for each unit region constituting the region where carbon-fixing organisms exist, a roughness that is an indicator of the density of carbon-fixing organisms present in the region where carbon-fixing organisms exist, based on the color of the carbon-fixing organisms and the color of the seabed, and also determines the distribution of the roughness for each unit region in the region where carbon-fixing organisms exist. The carbon-fixing organisms are organisms whose roughness varies depending on the time of year. The evaluation unit evaluates whether the amount of carbon fixed included in the application information matches the total amount of carbon fixed in the entire area determined from the distribution of roughness for each unit area, and evaluates the reliability taking this evaluation into account. The evaluation apparatus according to feature 1.

10. A roughness determination unit that determines roughness, which is an index indicating the density of carbon-fixing organisms in the region where the carbon-fixing organisms exist, by grasping the color information possessed by the carbon-fixing organisms from the types of carbon-fixing organisms captured in the reference image and detecting whether or not the grasped color information is present in the reference image, Based on the determined roughness, a fixation amount determination unit determines the amount of carbon fixed by the carbon-fixing organism in the region where the carbon-fixing organism is present, It comprises a carbon credit determination unit that determines the amount of carbon credits based on the determined amount of carbon sequestration. The evaluation apparatus according to feature 1.

11. The carbon credit determination unit further determines an amount from the determined carbon credit amount that is pre-associated with the applicant's transaction scale and carbon sequestration amount, and that corresponds to the determined carbon sequestration amount. The evaluation apparatus according to feature 10.

12. Performed by an evaluation device, A receiving step for receiving application information for an application regarding the amount of carbon fixed by the carbon-fixing organism, including a photograph of a carbon-fixing organism that fixes carbon to itself, the time the photograph was taken, and the position of the carbon-fixing organism in the photograph. The process includes a reference image of the carbon-fixing organism taken from above, which is an image taken at a time and location corresponding to the time the aforementioned image was taken, and an evaluation step of evaluating the reliability of the application information based on the aforementioned image, The evaluation step involves evaluating the reliability of the application information by assessing the consistency between the reference image and the captured image. The aforementioned consistency includes the degree of agreement between the reference image and the captured image, The degree of agreement includes the degree of agreement of the distribution of the outlines of the carbon-fixing organisms and the degree of agreement of the colors within the outlines for the reference image and the captured image. An evaluation method characterized by the following.

Images