Support systems, support methods, support devices, and programs

Abstract

This system provides support to ensure the accuracy of calculating decarbonization amounts. [Solution] The support system comprises a terminal equipped with a camera and a server capable of communicating with the terminal. The terminal captures an image of an object to be used for calculating the amount of decarbonization and transmits the image of the object to the server. The server comprises an acquisition unit that acquires the image of the object transmitted by the terminal and a recognition unit that recognizes features from the image of the object that are used to calculate the amount of decarbonization. The terminal displays the calculation result of the amount of decarbonization of the object based on the features.

Description

Technical Field

[0006] , , , ,

[0001] The present invention relates to a support system, a support method, a support device, and a program.

Background Art

[0002] In Patent Document 1, by mediating an exchange contract between a decarbonization point based on the amount of decarbonization, which is the amount of carbon dioxide emissions reduced by a user refraining from using automobiles and motorcycles and moving on foot or by bicycle, and a privilege provided by a client, a support system is disclosed that supports the maintenance of individual users' health and the realization of a decarbonized society.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in efforts towards decarbonization, there are various activities such as the use of portable water cylinders, the use of compost, and waste oil recovery. However, if the recording of these activities depends only on manual input or taking photos, there is a problem that it is difficult to ensure accuracy.

[0005] The present invention has been made in view of the above problems, and its main object is to provide a support system, a support method, a support device, and a program capable of ensuring the accuracy of calculating the amount of decarbonization.

Means for Solving the Problems

[0006] To solve the above problems, a support system according to one aspect of the present invention comprises a terminal equipped with a camera and a server capable of communicating with the terminal, wherein the terminal captures an image of an object to be used for calculating the amount of decarbonization and transmits the image of the object to the server, the server comprises an acquisition unit that acquires the image of the object transmitted by the terminal and a recognition unit that recognizes features from the image of the object that are used for calculating the amount of decarbonization, and the terminal displays the calculation result of the amount of decarbonization of the object based on the features.

[0007] Furthermore, in another embodiment of the present invention, a support method involves using a terminal equipped with a camera to capture an image of an object to be used for calculating the amount of decarbonization, recognizing features from the image of the object that are used for calculating the amount of decarbonization, calculating the amount of decarbonization of the object based on the features, and displaying the calculation result of the amount of decarbonization on the terminal.

[0008] Furthermore, another embodiment of the present invention provides a support device comprising: an acquisition unit that acquires an image of an object to be used for calculating the amount of decarbonization, captured by a terminal equipped with a camera; and a relay unit that causes a recognition unit to recognize features used for calculating the amount of decarbonization from the image of the object.

[0009] Furthermore, a program in another aspect of the present invention causes a computer to perform the following actions: acquire an image of an object to be used for calculating the amount of decarbonization, which is captured by a terminal equipped with a camera; and cause a recognition unit to recognize features from the image of the object that are used for calculating the amount of decarbonization. [Effects of the Invention]

[0010] According to the present invention, it is possible to ensure the accuracy of calculating the amount of decarbonization. [Brief explanation of the drawing]

[0011] [Figure 1] This is a diagram showing an example of a support system. [Figure 2] This figure shows an example of a support device. [Figure 3] This is a diagram showing an example of a terminal. [Figure 4]It is a diagram showing an example of an imaging screen of a terminal. [Figure 5] It is a diagram showing an example of a notification screen of a terminal. [Figure 6] It is a diagram showing an example of a support device. [Figure 7] It is a diagram explaining the learning phase. [Figure 8] It is a diagram explaining the inference phase. [Figure 9] It is a diagram showing an example of a support method. [Figure 10] It is a diagram showing an example of an imaging screen of a terminal. [Figure 11] It is a diagram showing an example of an imaging screen of a terminal. [Figure 12] It is a diagram showing an example of a support method.

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this specification and each figure, elements that are the same as those described above with respect to the already shown figures may be assigned the same reference numerals, and detailed descriptions may be omitted as appropriate.

[0013] FIG. 1 is a diagram showing an example of a support system 100. The support system 100 includes a server 1 and a terminal 2 possessed by a user UZ. The server 1 and the terminal 2 are connected to a communication network NT such as the Internet and can communicate with each other via the network.

[0014] The server 1 is composed of a plurality of server computers including a support device 3 and an AI (Artificial Intelligence) server 4. However, the server 1 is not limited to this and may be composed of one server computer.

[0015] The terminal 2 is a portable wireless information terminal such as a smartphone or a tablet computer.

[0016] The support system 100 is a system for supporting the behavioral transformation of user UZ towards decarbonization by visualizing the decarbonization amount of object BJ related to the decarbonization actions of user UZ.

[0017] Object BJ is, for example, a bottle, food waste, a lunch box container, a container containing waste oil, food, an invoice, and fresh food, etc. The decarbonization amount is an amount representing the reduction in the emission amount of greenhouse gases such as carbon dioxide.

[0018] Figure 2 is a block diagram showing a configuration example of the support device 3. The support device 3 is a computer including a processing unit 11, a storage unit 12, and a communication unit 13. The AI server 4 also has a similar configuration.

[0019] The processing unit 11 includes a CPU and executes information processing according to a program stored in the storage unit 12. The storage unit 12 includes a RAM, a ROM, and a non-volatile memory, etc. The communication unit 13 is an interface for connecting the support device 3 to the communication network NT.

[0020] The program may be provided to the support device 3 via a non-temporary storage medium or may be provided to the support device 3 via a communication line.

[0021] Figure 3 is a block diagram showing a configuration example of the terminal 2. The terminal 2 is also a computer including a processing unit 21, a storage unit 22, and a communication unit 23, similar to the support device 3. The terminal 2 further includes a touch panel 24, a camera 25, and a GNSS receiver 26.

[0022] In addition to the communication unit 23 that realizes a mobile communication or wireless LAN communication method for connecting the terminal 2 to the communication network NT, the communication unit 23 further includes a communication unit that realizes short-range wireless communication such as Bluetooth (registered trademark).

[0023] The touch panel 24 integrates the display unit and the operation unit. The GNSS receiver 26 detects the position of the terminal 2 based on radio waves received from the GNSS (Global Navigation Satellite System).

[0024] The processing unit 21, according to the application program stored in the memory unit 22, captures an image of the target object BJ using the camera 25 and transmits the image of the target object BJ to the server 1. As shown in Figure 4, the terminal 2 displays the imaging screen PS for capturing the target object BJ.

[0025] The imaging screen PS has a display area SA where the object BJ being captured by camera 25 is displayed, an imaging button BT, and a string of characters CL prompting the user to take an image. When the user UZ operates the imaging button BT, an image of the object BJ is captured.

[0026] Furthermore, the processing unit 21 receives a response from the server 1 and displays the calculation result of the decarbonization amount based on the characteristics of the target object BJ. As shown in Figure 5, the terminal 2 displays a notification screen NS that shows the calculation result DC of the decarbonization amount.

[0027] Figure 6 is a block diagram showing an example of the functional configuration of the support device 3 and the AI ​​server 4.

[0028] The support device 3 comprises an acquisition unit 31, a relay unit 32, a calculation unit 33, and a notification unit 34. These functional units 31 to 34 are realized by the processing unit 11 executing information processing according to a program. The support device 3 also comprises an image storage unit 39. The image storage unit 39 may be built into the storage unit 12 or into an external storage device.

[0029] The AI ​​server 4 includes a recognition unit 41. This function unit 41 is realized by the processing unit of the AI ​​server 4 executing information processing according to a program. The AI ​​server 4 also includes a model storage unit 49. The model storage unit 49 may be built in the storage unit of the AI ​​server 4, or it may be built in an external storage device.

[0030] The acquisition unit 31 of the support device 3 acquires an image of the target object BJ transmitted by the terminal 2 and stores it in the image storage unit 39.

[0031] The relay unit 32 causes the recognition unit 41 of the AI ​​server 4 to recognize features from the image of the target object BJ that will be used to calculate the amount of carbon decarbonization.

[0032] The calculation unit 33 calculates the amount of decarbonization of the object BJ based on the features recognized by the recognition unit 41 of the AI ​​server 4.

[0033] The notification unit 34 notifies terminal 2 of the calculation result of the amount of decarbonization, and causes terminal 2 to display the calculation result of the amount of decarbonization.

[0034] The entity implementing each functional unit is not limited to the examples above. For example, the support device 3 may be equipped with a recognition unit 41 to recognize features from an image of the target object BJ. Alternatively, the terminal 2 may be equipped with a calculation unit 33 to calculate the amount of decarbonization based on the features acquired from the support device 3.

[0035] The recognition unit 41 of the AI ​​server 4 recognizes features from the image of the object BJ that are used to calculate the amount of carbon decarbonization.

[0036] The recognition unit 41 recognizes features from the image of the object BJ using a trained model stored in the model storage unit 49. However, the recognition unit 41 may also recognize features from the image of the object BJ using rule-based image recognition.

[0037] The model memory unit 49 stores multiple trained models prepared for each type of object BJ, and the recognition unit 41 uses the trained model corresponding to the type of object BJ to recognize features from the image of the object BJ.

[0038] The characteristics used to calculate the amount of decarbonization include, for example, characteristics related to the quantity (volume, weight, or size) of the object BJ. The amount of decarbonization of the object BJ is calculated according to the recognized quantity.

[0039] Furthermore, the characteristics used in calculating the amount of decarbonization may include the suitability of the object BJ. The amount of decarbonization of the object BJ is calculated if its suitability is above a specified level, and not calculated if its suitability is below a specified level.

[0040] The following is an overview of the training and inference phases of the trained model.

[0041] As shown in Figure 7, in the learning phase, machine learning is performed on the model MD (the model before training) using the training images MG and the training data for quantity and qualification. The model MD is an image recognition model such as a convolutional neural network (CNN).

[0042] One element of the output layer of Model MD is composed of, for example, an identity function and outputs a numerical value representing the "quantity". Another element of the output layer of Model MD is composed of, for example, a sigmoid function and outputs a numerical value between 0 and 1 that represents the "eligibility".

[0043] The "quantity" of the training data is a numerical value representing capacity, weight, or volume. The "eligibility" of the training data is a binary label indicating eligibility or non-eligibility. For example, the OK label, indicating eligibility, is represented by 1, and the NG label, indicating non-eligibility, is represented by 0.

[0044] In machine learning, training images (MG) are input to the model (MD), the difference between the quantity and accuracy output from the model (MD) and the quantity and accuracy of the training data is calculated, and the parameters of the model (MD) are adjusted by performing backpropagation to reduce the calculated difference.

[0045] The trained model LM generated in this way is stored in the model storage unit 49.

[0046] As shown in Figure 8, during the inference phase, the recognition unit 41 inputs the image SG of the object BJ transmitted by the terminal 2 into the trained model LM and outputs a numerical value representing the "quantity" and a numerical value representing the "qualification".

[0047] Specific examples of generating pre-trained models for each type of object BJ, recognizing features, and calculating the amount of carbon decarbonization will be described in detail later.

[0048] Figure 9 is a flowchart showing an example of the procedure for the support method implemented in the support system 100. The processing unit 21 of terminal 2 executes processes S21 to S24 shown on the left side of the figure according to the program. The processing unit 11 of support device 3 executes processes S31 to S34 shown in the center of the figure according to the program. The processing unit of AI server 4 executes processes S41 to S42 shown on the right side of the figure according to the program.

[0049] First, terminal 2 uses camera 25 to image the target object BJ according to the user UZ's instructions (S21). The imaging screen PS (see Figure 4) is used for imaging. An imaging screen PS is prepared for each type of target object BJ and is selected by the user UZ before imaging.

[0050] Next, terminal 2 assigns a tag representing the type of object BJ to the image of the captured object BJ (S22). Here, a tag corresponding to the imaging screen PS used to capture the object BJ is selected and assigned to the image.

[0051] Next, terminal 2 transmits an image of the tagged object BJ to the support device 3 (S23).

[0052] For example, if the object BJ is a bottle, user UZ uses the bottle-specific imaging screen PS to image the bottle. As a result, the image of the bottle is tagged to represent the bottle.

[0053] The support device 3 acquires an image of the target object BJ transmitted by the terminal 2 (S31, processing as the acquisition unit 31).

[0054] Next, the support device 3 relays the image of the target object BJ (S32, processing as relay unit 32). That is, the support device 3 causes the AI ​​server 4 to recognize features from the image of the target object BJ that will be used to calculate the amount of decarbonization.

[0055] The AI ​​server 4 recognizes features used to calculate the amount of decarbonization from the image of the target object BJ transmitted by the support device 3 (S41, processing as the recognition unit 41), and transmits the recognized features to the support device 3 (S42).

[0056] At this time, the support device 3 specifies that it should use a pre-trained model corresponding to the tag attached to the image of the object BJ, and the AI ​​server 4 recognizes the features using the specified pre-trained model. The selection of the pre-trained model corresponding to the tag may also be performed by the AI ​​server 4.

[0057] For example, if the object BJ is a bottle, the bottle's features are recognized using a pre-trained model for bottles that corresponds to the tag representing the bottle attached to the image.

[0058] The support device 3 calculates the amount of decarbonization of the target object BJ based on the characteristics transmitted by the AI ​​server 4 (S33, processing as calculation unit 33), and notifies the terminal 2 of the calculation result of the amount of decarbonization (S34, processing as notification unit 34).

[0059] Terminal 2 displays the calculation result of the decarbonization amount notified by the support device 3 (S24). The notification screen NS (see Figure 5) is used to display the decarbonization amount. Note that the calculation of the decarbonization amount may also be performed on terminal 2.

[0060] As explained above, by visualizing the amount of carbon decarbonization of the target object BJ, users UZ can visually understand the object BJ's contribution to carbon decarbonization, thereby encouraging behavioral changes toward effective carbon decarbonization actions. Furthermore, by calculating the amount of carbon decarbonization after recognizing the characteristics of the target object BJ, accuracy can be ensured. In other words, it is possible to avoid misunderstandings, incorrect inputs, and false inputs that may occur when relying solely on manual input or photographs, thereby ensuring the accuracy of the carbon decarbonization calculation.

[0061] Furthermore, by awarding decarbonization points to users (UZ) based on their decarbonization efforts and allowing them to exchange these points for rewards offered within communities such as local governments, it becomes possible to increase the motivation of users (UZ) to accumulate decarbonization points and further encourage behavioral change among them.

[0062] The following sections describe specific examples of generating pre-trained models for each type of object BJ, recognizing features, and calculating the amount of carbon decarbonization.

[0063] (1) My bottle The manufacturing, disposal, and recycling of plastic bottles release a significant amount of carbon dioxide. Therefore, choosing not to buy bottled beverages and instead carrying a reusable water bottle (a so-called "my bottle") is a decarbonization action. Therefore, in the example where the object BJ is a bottle, the capacity of the bottle and its suitability as a portable water bottle are recognized by a trained model, and the amount of carbon decarbonization achieved by using the bottle as a portable water bottle is calculated. For example, by assigning an "OK" label to training images of portable bottles with lids made of stainless steel, aluminum, plastic, or silicone, and an "NG" label to training images of containers without lids made of paper, plastic, stainless steel, ceramic, or glass, it becomes possible to determine whether an item is suitable as a portable water bottle by generating a trained model. The amount of decarbonization is calculated based on the bottle's capacity, provided the bottle is suitable as a portable water bottle; it is not calculated if the bottle is not suitable as a portable water bottle.

[0064] (2) Compost Because food waste contains a lot of moisture, disposing of it as combustible waste releases a large amount of carbon dioxide during the incineration process. Therefore, composting food waste is a decarbonization action. Therefore, in the example where the target object BJ is food waste, the weight or volume of the food waste and its suitability for composting are recognized by a trained model, and the amount of decarbonization achieved by processing the food waste in compost is calculated. For example, by assigning an "OK" label to training images of compostable waste (especially those with a soil background, outdoors, or a sink), and an "NG" label to training images of waste containing inappropriate materials, waste in a trash can, or non-combustible waste, it becomes possible to generate a trained model that can determine suitability for composting. The amount of decarbonization is calculated based on weight or volume if the food waste is suitable for composting, but not if it is not suitable for composting.

[0065] (3) Lunch box container The manufacturing, disposal, and recycling of disposable plastic or paper lunch containers release a significant amount of carbon dioxide. Therefore, bringing your own lunch in a reusable container is a decarbonization action. Therefore, in the example where the object BJ is a lunchbox container, the trained model recognizes whether the lunchbox container qualifies as waste, and calculates the amount of decarbonization achieved by using a lunchbox container that does not qualify as waste. For example, by assigning an "OK" label to training images of non-disposable plastic or metal lunch containers and an "NG" label to training images of disposable plastic or paper containers (so-called lunch container casings), and generating a trained model, it becomes possible to determine whether or not a lunch container qualifies as waste. The amount of decarbonization is calculated only when the lunch box container does not qualify as waste (i.e., it is not disposable), and not when the lunch box container does qualify as waste. The amount of decarbonization may be calculated according to the capacity of the lunch box container, or it may be a uniform amount.

[0066] (4) Waste oil Disposing of used cooking oil at home can lead to environmental pollution if poured down the drain, and incineration releases a large amount of carbon dioxide if it is thrown away as waste. Therefore, taking used cooking oil to a collection point is a decarbonization action. Therefore, in the example where the object BJ is a container of waste oil, the capacity of the container and its suitability for waste oil collection are recognized by a trained model, and the amount of decarbonization achieved through waste oil collection is calculated. For example, by assigning an "OK" label to training images of brown or yellow oil in a dedicated container or a transparent container such as a plastic bottle (especially those showing a collection spot in the background), and assigning an "NG" label to training images of oil mixed with food scraps or other liquids, oil with a lot of dirt or solid matter, machine oil, other liquids such as beverages, or oil that has been solidified or absorbed by paper, it becomes possible to generate a trained model and determine the suitability of the waste oil for collection. The amount of decarbonization is calculated based on the volume of the container if the container containing the waste oil is suitable for collection, but not if the container is not suitable for collection.

[0067] (5) Food drive Donating surplus food from your home to welfare organizations instead of throwing it away is known as a food drive. Donating food instead of throwing it away is also a decarbonization action. Therefore, in cases where the target object BJ is food, the eligibility of the food as donated food is recognized by a trained model, and the amount of carbon decarbonization achieved by donating the food to a food drive is calculated. For example, by assigning an "OK" label to training images of foods suitable for donation, such as rice, dried noodles, canned goods, retort foods, or instant foods, and assigning an "NG" label to training images of foods unsuitable for donation, such as opened food, fresh food, or food that is past its expiration date or has an unknown expiration date, it becomes possible to determine the eligibility of food as a donation. The amount of decarbonization is calculated only if the food is suitable for donation, and not if the food is not suitable for donation. The amount of decarbonization may be calculated based on the weight or volume of the food, or it may be a fixed amount. The application on terminal 2 displays food drive collection spots on a map and detects the implementation of a food drive when a donation operation is performed at one of the collection spots. The detection of the implementation of a food drive may also be performed by the support device 3 after it has acquired the location of terminal 2.

[0068] (6) Meter reading slip Reducing the use of electricity or gas, etc., constitutes a decarbonization action. Therefore, in the example where the target object BJ is a meter reading slip, the trained model recognizes the usage amount for the target period and the usage amount for past periods as recorded on the meter reading slip, and calculates the amount of decarbonization based on the difference between the usage amount for the target period and the usage amount for past periods. For example, as shown in Figure 10, the imaging screen PS for the meter reading slip contains the string CU indicating this month's usage and the string PU indicating last month's usage. From these, the current month's usage and last month's usage are recognized, the decrease in this month's usage compared to last month's usage is calculated, and the amount of decarbonization corresponding to the decrease is calculated. The target period and past period are not limited to months, but may also be in years.

[0069] (7)Fresh food Consuming fresh food produced locally within the same region or a neighboring region (so-called local production and consumption) is a decarbonization action because it is expected to reduce carbon dioxide emissions related to transportation. Therefore, in the example where the target object BJ is fresh food, the origin of the fresh food is recognized from the label attached to the fresh food using a trained model, and the amount of decarbonization is calculated based on a comparison between the location of terminal 2 and the origin of the fresh food. For example, as shown in the imaging screen PS for fresh food in Figure 11, fresh food is attached to a label LB, and information such as the place of origin and weight is recognized from the label LB by a trained model. Furthermore, for example, by assigning OK labels to training images of individually labeled fresh produce and NG labels to training images of fresh produce sold in baskets including price tags, it becomes possible to determine the suitability of labels by generating a trained model. Furthermore, by assigning an "OK" label to training images that show the interior of a house in the background, and an "NG" label to training images that show the interior of a store in the background, and generating a trained model, it becomes possible to determine the suitability of the imaging location. In other words, it becomes possible to determine whether the fresh food is unpurchased or has already been purchased.

[0070] Figure 12 is a flowchart showing an example of a procedure for providing support when the target object BJ is a perishable food item. For steps corresponding to the flowchart in Figure 9 above, the same number is used, and detailed explanations are omitted.

[0071] After purchasing fresh food, user UZ launches the application program on terminal 2 at home or elsewhere and uses the imaging screen PS (see Figure 11) to capture images of the fresh food with camera 25 (S21).

[0072] Next, terminal 2 obtains its position using the GNSS receiver 26 (S26), adds a tag representing fresh food and the position of terminal 2 to the image (S22), and transmits the image to the support device 3 (S23).

[0073] The AI ​​server 4 recognizes information such as the place of origin and weight indicated on the label LB from the image transmitted by the support device 3 (S41) and transmits it to the support device 3 (S42).

[0074] The support device 3 calculates the amount of decarbonization based on a comparison between the location of terminal 2 and the place of origin of the fresh food (S33). For example, if the place of origin of the fresh food is the same as or adjacent to the location of terminal 2, the support device 3 calculates the amount of decarbonization according to the weight of the fresh food. For the weight of the fresh food, for example, the general weight per unit quantity of the target food, as described in the "Food Composition Table" included in the label LB, is used.

[0075] The decarbonization amount for fresh foods is calculated as the extent to which carbon dioxide emissions during the distribution of fresh foods are reduced compared to the general emissions for similar fresh foods. The calculation of decarbonization uses a default value for transport emissions per unit weight.

[0076] For example, if you purchase tomatoes from Fukuoka Prefecture in Fukuoka, the amount of decarbonization will be calculated using the carbon dioxide emissions from the transportation of tomatoes commonly sold in local supermarkets (for example, Hiroshima Prefecture) as a comparative value. There are various ways to determine the comparative value; for example, the carbon dioxide emissions from the transportation of the main production area of ​​the fresh food in question (for example, Ehime or Wakayama for mandarins) could be used as the comparative value.

[0077] Once data is read from the label LB of fresh food, it is cross-referenced with data such as origin, producer, and expiration date, and treated as used, so that the amount of decarbonization is not calculated twice for the same food product.

[0078] In this example, fresh food was imaged using camera 25, and the production area and weight were recognized from the label LB included in the image to calculate the amount of decarbonization. However, the method is not limited to this, and may be done as follows.

[0079] For example, the terminal 2 may read code data from a code image representing a two-dimensional code such as a QR code or a one-dimensional code such as a barcode printed on label LB, extract the place of production and weight from the code data, and calculate the amount of decarbonization.

[0080] Furthermore, if the label LB is an electronic tag, the code data may be read from the label LB using the NFC (Near Field Communication) or other short-range wireless communication function of terminal 2, and the production location and weight may be extracted from the code data to calculate the amount of decarbonization.

[0081] Alternatively, the location of terminal 2 may be compared with the location of a pre-registered store, and the amount of decarbonization of fresh food may be calculated when terminal 2 is outside the store, but the amount of decarbonization may not be calculated when terminal 2 is outside the store.

[0082] Furthermore, the conditions for awarding and using decarbonization points, which are granted based on the amount of decarbonization achieved, may be varied on a community-by-community basis. For example, decarbonization points related to local production and consumption, such as those awarded based on the amount of decarbonization achieved for fresh food, as in this example, may be limited to use only within the local community, or the region may be given preferential treatment over other regions in terms of the amount of decarbonization points awarded. This can further strengthen user behavioral changes toward decarbonization actions related to local production and consumption.

[0083] Furthermore, as mentioned above, fresh produce sold in baskets was deemed unacceptable because purchase cannot be verified. However, for example, one could take a picture of the fresh produce sold in baskets and its label at the store, then obtain and photograph the corresponding receipt, and then compare these two images to treat it as a purchased item.

[0084] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above, and various modifications are of course possible for those skilled in the art.

[0085] The following are some representative embodiments.

[0086] In a first embodiment, the support system comprises a terminal equipped with a camera and a server capable of communicating with the terminal, wherein the terminal captures an image of an object to be used for calculating the amount of decarbonization and transmits the image of the object to the server, the server comprises an acquisition unit that acquires the image of the object transmitted by the terminal and a recognition unit that recognizes features from the image of the object that are used for calculating the amount of decarbonization, and the terminal displays the calculation result of the amount of decarbonization of the object based on the features. This makes it possible to ensure the accuracy of the calculation of the amount of decarbonization.

[0087] In a second embodiment, in the first embodiment, the recognition unit recognizes the characteristics relating to the quantity of the object, and the amount of decarbonization may be calculated according to the quantity. This makes it possible to ensure the accuracy of calculating the amount of decarbonization by utilizing the quantity of the object.

[0088] In a third embodiment, in the first embodiment, the recognition unit recognizes the suitability of the object as a characteristic, and the amount of decarbonization does not need to be calculated if the suitability is below a predetermined level. This makes it possible to ensure the accuracy of calculating the amount of decarbonization by utilizing the suitability of the object.

[0089] In a fourth embodiment, in the first embodiment, the object is a bottle, the recognition unit recognizes the capacity of the bottle and its suitability as a portable water bottle, and the amount of decarbonization may be the amount of decarbonization achieved by using the bottle as a portable water bottle. This makes it possible to ensure the accuracy of calculating the amount of decarbonization achieved by using a portable water bottle.

[0090] In a fifth embodiment, in the first embodiment, the object is food waste, the recognition unit recognizes the weight or volume of the food waste and its suitability for composting, and the amount of decarbonization may be the amount of decarbonization achieved by processing the food waste in the compost. This makes it possible to ensure the accuracy of calculating the amount of decarbonization achieved by processing food waste in the compost.

[0091] In a sixth embodiment, in the first embodiment, the object is a lunch box container, the recognition unit recognizes whether the lunch box container is waste, and the amount of decarbonization may be the amount of decarbonization achieved by using the lunch box container which is not considered waste. This makes it possible to ensure the accuracy of calculating the amount of decarbonization achieved by using the lunch box container which is not considered waste.

[0092] In a seventh embodiment, in the first embodiment, the object is a container containing waste oil, the recognition unit recognizes the capacity of the container and its suitability for waste oil recovery, and the decarbonization amount may be the amount of decarbonization due to the recovery of the waste oil. This makes it possible to ensure the accuracy of calculating the amount of decarbonization due to waste oil recovery.

[0093] In the eighth embodiment, in the first embodiment, the object is food, the recognition unit recognizes the eligibility of the food as donated food, and the amount of decarbonization may be the amount of decarbonization achieved by donating the food to a food drive. This makes it possible to ensure the accuracy of calculating the amount of decarbonization achieved by donating food to a food drive.

[0094] In the ninth embodiment, in the first embodiment, the object is a meter reading slip, the recognition unit recognizes the amount of usage for the target period and the amount of usage for past periods as indicated on the meter reading slip, and the amount of decarbonization may be the amount of decarbonization corresponding to the difference between the amount of usage for the target period and the amount of usage for past periods. This makes it possible to ensure the accuracy of calculating the amount of decarbonization in response to changes in usage.

[0095] In a tenth embodiment, in the first embodiment, the terminal is equipped with an acquisition unit that acquires location data of the terminal, the object is fresh food, the recognition unit recognizes the place of origin of the fresh food from a label attached to the fresh food, and the amount of decarbonization may be an amount of decarbonization based on a comparison between the location of the terminal and the place of origin of the fresh food. This makes it possible to ensure the accuracy of calculating the amount of decarbonization based on the place of origin of the fresh food.

[0096] In an eleventh embodiment, in the first embodiment, the terminal may assign a tag representing the type of object to the image of the object, and the recognition unit may recognize the features using a model corresponding to the tag from among a plurality of models prepared for each type of object. This makes it possible to ensure the accuracy of calculating the amount of decarbonization for each type of object.

[0097] In the twelfth embodiment, the support method involves using a terminal equipped with a camera to capture an image of an object to be used for calculating the amount of decarbonization, recognizing features from the image of the object that are used for calculating the amount of decarbonization, calculating the amount of decarbonization of the object based on these features, and displaying the calculation result of the amount of decarbonization on the terminal. This makes it possible to ensure the accuracy of the calculation of the amount of decarbonization.

[0098] In a thirteenth embodiment, the support device comprises an acquisition unit that acquires an image of an object to be used for calculating the amount of decarbonization, captured by a terminal equipped with a camera, and a relay unit that causes a recognition unit to recognize features used for calculating the amount of decarbonization from the image of the object. This makes it possible to ensure the accuracy of the calculation of the amount of decarbonization.

[0099] In the 14th embodiment, the program causes a computer to acquire an image of the object to be used for calculating the amount of decarbonization, which is captured by a terminal equipped with a camera, and to cause a recognition unit to recognize features from the image of the object that will be used for calculating the amount of decarbonization. This makes it possible to ensure the accuracy of the calculation of the amount of decarbonization. [Explanation of Symbols]

[0100] 1 Server, 2 Terminal, 3 Support device, 4 AI server, 11 Processing unit, 12 Storage unit, 13 Communication unit, 21 Processing unit, 22 Storage unit, 23 Communication unit, 24 Touch panel, 25 Camera, 26 GNSS receiver, 31 Acquisition unit, 32 Relay unit, 33 Calculation unit, 34 Notification unit, 39 Image storage unit, 41 Recognition unit, 49 Model storage unit, 100 Support system

Claims

1. A device equipped with a camera, A server that can communicate with the aforementioned terminal, Equipped with, The terminal captures an image of the object to be used for calculating the amount of decarbonization, and transmits the image of the object to the server. The aforementioned server, An acquisition unit that acquires an image of the object transmitted by the terminal, A recognition unit that recognizes features used to calculate the amount of decarbonization from an image of the object, Equipped with, The terminal displays the calculation result of the amount of decarbonization of the object based on the characteristics described above. Support system.

2. The recognition unit recognizes the characteristics relating to the quantity of the object, The amount of decarbonization is calculated according to the amount. The support system according to claim 1.

3. The recognition unit recognizes the suitability of the object as the characteristic, The amount of decarbonization mentioned above is not calculated if the eligibility is below a predetermined level. The support system according to claim 1.

4. The aforementioned object is a bottle, The recognition unit recognizes the capacity of the bottle and the suitability of the bottle as a portable water bottle. The amount of decarbonization mentioned above is the amount of decarbonization achieved by using the bottle as a portable water bottle. The support system according to claim 1.

5. The aforementioned object is food waste, The recognition unit recognizes the weight or volume of the food waste and the suitability of the food waste for composting. The amount of decarbonization mentioned above is the amount of decarbonization achieved by processing the food waste in the compost. The support system according to claim 1.

6. The aforementioned object is a lunch box container. The recognition unit recognizes that the lunch box container is waste, The aforementioned decarbonization amount is the amount of decarbonization achieved by using the lunch box containers that do not fall under the category of waste. The support system according to claim 1.

7. The aforementioned object is a container containing waste oil. The recognition unit recognizes the capacity of the container and its suitability for the collection of the waste oil. The aforementioned decarbonization amount is the amount of decarbonization achieved through the recovery of the waste oil. The support system according to claim 1.

8. The aforementioned object is food, The recognition unit recognizes the eligibility of the food as donated food, The aforementioned amount of decarbonization is the amount of decarbonization achieved by donating the food to a food drive. The support system according to claim 1.

9. The aforementioned object is a meter reading slip, The recognition unit recognizes the amount of usage for the target period and the amount of usage for past periods as stated on the meter reading slip. The amount of decarbonization is the amount of decarbonization corresponding to the difference between the amount used during the target period and the amount used during the past period. The support system according to claim 1.

10. The terminal is equipped with an acquisition unit that acquires location data of the terminal, The aforementioned object is fresh food, The recognition unit recognizes the origin of the fresh food from the label attached to the fresh food, The amount of decarbonization is the amount of decarbonization based on a comparison between the location of the terminal and the origin of the fresh food. The support system according to claim 1.

11. The terminal assigns a tag to the image of the object that indicates the type of object, The recognition unit recognizes the features using a model corresponding to the tag from among a plurality of models prepared for each type of object. The support system according to claim 1.

12. A terminal equipped with a camera will take an image of the object to be used for calculating the amount of decarbonization, From the image of the object, the features used to calculate the amount of decarbonization are recognized. Based on the above characteristics, the amount of decarbonization of the object is calculated, The calculation result of the amount of decarbonization is displayed on the terminal. A method of assisting a computer in performing a task.

13. An acquisition unit that acquires images of objects to be used for calculating the amount of decarbonization, captured by a terminal equipped with a camera, The recognition unit includes a relay unit that recognizes features from the image of the object that are used to calculate the amount of decarbonization, A support device equipped with the following features.

14. Acquiring images of objects to be used for calculating decarbonization amount, captured by a terminal equipped with a camera, and The recognition unit is instructed to recognize features from the image of the object that are used to calculate the amount of decarbonization. A program that causes a computer to execute something.

Images