Recovery support device, recovery system, recovery support method, and program

Abstract

The goal is to minimize the burden on users involved in resource collection while utilizing the information obtained through collection. [Solution] The system is installed at a collection point of a recycling company that recycles materials including packaging materials, and includes an acquisition unit that acquires target images of the materials collected at the collection point and transported to the collection point by the recycling company, a determination unit that determines the category corresponding to the materials based on the target images, and a summation unit that aggregates the amount of materials collected for each category determined by the determination unit.

Description

Technical Field

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

Background Art

[0002] Some packaging materials for products such as food and detergents are made of recyclable materials. For such packaging materials, after the product is used by the user, it can be recycled as a resource by being collected. Systems for efficiently recycling packaging materials have also been devised (see, for example, Patent Document 1). In this system, by having a collection device placed at the storefront of a store such as a supermarket or within a local government facility read individual identification information such as a JAN code printed on the packaging material, the packaging materials are sorted and collected by material.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above system, since the user has to go to the collection device placed at the storefront or the like, the burden is large and it is difficult to improve the collection rate. On the other hand, companies that manufacture products eligible for recycling are required to take part in recycling efforts toward a circular economy. For example, some companies publish their recycling targets on their websites and demonstrate their commitment to achieving those targets. For such companies, obtaining data from recycling companies showing the amount of recyclable products collected would be useful as it would objectively demonstrate that recycling is taking place. Furthermore, operators of recycling infrastructure can expect to gain cooperation from companies that provide recyclable products by providing them with useful information, such as having them bear the operating costs of recycling. Therefore, if the information obtained through collection can be utilized while minimizing the burden on users, it would be desirable to provide useful information to companies and to make effective use of resources.

[0005] This invention has been made in view of these circumstances, and its purpose is to provide a resource recovery support device, a resource recovery system, a resource recovery support method, and a program that can utilize information obtained through resource recovery while reducing the burden on users involved in resource recovery. [Means for solving the problem]

[0006] To solve the above-mentioned problems, one aspect of the present invention is a collection support device installed at a collection center of a recycling company that recycles collection targets including packaging materials, and comprising: an acquisition unit that acquires a target image G of the collection targets that have been collected at the collection center and transported to the collection center by the recycling company; a determination unit that determines the category corresponding to the collection targets based on the target image; and a summation unit that aggregates the amount of collection for each category determined by the determination unit.

[0007] Furthermore, one aspect of the present invention is a recovery system comprising the recovery support device described above and a sorting device that sorts according to the categories determined by the determination unit.

[0008] Furthermore, one aspect of the present invention is a collection support method performed by a collection support device, which is a computer installed at a collection center of a recycling company that recycles collected materials including packaging materials, the method comprising acquiring target images of the collected materials that have been collected at the collection center and transported to the collection center by the recycling company, determining the category corresponding to the collected materials based on the target images, and aggregating the amount of materials collected for each determined category.

[0009] Furthermore, one aspect of the present invention is a program to be executed on a computer installed at a collection point of a recycling company that recycles collected materials including packaging materials, the program includes acquiring target images of the collected materials that have been collected at the collection point and transported to the collection point by the recycling company, determining the category corresponding to the collected materials based on the target images, and aggregating the amount of each determined category. [Effects of the Invention]

[0010] As explained above, this invention makes it possible to utilize the information obtained through resource recovery while reducing the burden on users related to resource recovery. [Brief explanation of the drawing]

[0011] [Figure 1] This is a diagram showing the configuration of a recovery system 1 according to one embodiment of the present invention. [Figure 2] This is a block diagram showing the configuration of the recovery support device 10. [Figure 3A] This figure shows an example of the target image G. [Figure 3B] This figure shows another example of the target image G. [Figure 4] This figure shows an example of information stored in the sorting information storage unit 106. [Figure 5] This figure shows an example of the information stored in the aggregated information storage unit 107. [Figure 6] This is a sequence diagram showing the flow of the recovery process performed by recovery system 1. [Figure 7] This flowchart shows the sorting process performed by the recovery support device 10. [Figure 8] This figure shows another example of information stored in the sorting information storage unit 106. [Figure 9] This figure shows another example of information stored in the aggregated information storage unit 107. [Figure 10] This is a flowchart showing the processing flow performed by the recovery support device 10. [Modes for carrying out the invention]

[0012] (First embodiment) Figure 1 is a diagram showing the configuration of a recovery system 1 according to one embodiment of the present invention. The recovery system 1 comprises a recovery support device 10, a sorting plant WS, and a manufacturer server 20. The recovery support device 10 and the manufacturer server 20 are connected via a communication network NW. The recovery support device 10 is connected via communication to a camera C and a sorting device 11 installed in the sorting plant WS. The recovery support device 10 may be installed in the sorting plant WS, or it may be installed in a different facility (for example, a management facility in the recovery system).

[0013] The items to be collected T (items T1, T2, etc.) are supplied to the sorting plant WS. The items to be collected T are used recyclable products, etc., that are brought to collection points by local governments and other locations for collection. For example, they are collected in collection bags P by collection vehicles V managed by collection companies 30 and transported to the sorting plant WS. Items to be collected (T) are, for example, articles made from recyclable materials (e.g., plastic). Examples of such articles include packaging materials, containers, cosmetics, stationery, electrical appliances, and consumables. The object to be recovered T may be, for example, something recovered in accordance with the resource recovery of plastics in a local government, etc., or may include something recovered by a product manufacturer corresponding to the object to be recovered T, for example, a cosmetics manufacturer or a beverage manufacturer, etc., by implementing its own recycling program. Further, it may include something sorted from general waste such as plastic waste (plastic garbage) and sent for resource recovery, etc. Examples of packaging materials include packages for food and beverages, containers containing toiletry products, containers containing refill detergents, etc. As the material, plastic may be partially used. Further, the objects to be recovered may include, in addition to articles made of recyclable materials, articles (such as products) targeted for reuse, recall, self-recovery, recovery of used items, etc. The container is a container for containing food or beverages, and as the material, plastic may be used partially or entirely.

[0014] The sorting factory WS is provided with a camera C, a sorting device 11, and a belt conveyor BC. The object to be recovered T is taken out of the recovery bag P by an operator or the like and placed on the upper surface of the upper belt of the belt conveyor BC. The object to be recovered T placed on the belt conveyor BC is conveyed in the conveying direction D. The camera C is provided above the belt conveyor BC. The camera C images the conveyed object to be recovered T from above and outputs an image (target image G) of the object to be recovered T to the recovery support device 10.

[0015] The recovery support device 10 is a computer that supports the recovery of the object to be recovered T. As the recovery support device 10, a PC (personal computer), a server, a cloud, etc. can be applied. The recovery support device 10 acquires the target image G from the camera C and generates sorting information based on the acquired target image G. The sorting information is information associating the object to be recovered T with the sorting destination, and is at least information associating whether or not the object to be recovered T is to be a sorting object for each sorting object. The sorting object is an object to be recycled as a resource, for example, a PET bottle, a resource plastic, etc. For example, if the object to be sorted is a PET bottle, the collection support device 10 determines whether the object to be collected T is a PET bottle based on the target image G, and generates sorting information that associates the information indicating the determination result with the object to be collected T. If the object to be sorted is a recyclable plastic, the collection support device 10 determines whether the object to be collected T is a recyclable plastic based on the target image G, and generates sorting information that associates the information indicating the determination result with the object to be collected T. The method by which the collection support device 10 determines whether the object to be collected T is an object to be sorted based on the target image G will be explained in detail later. The collection support device 10 outputs the generated sorting information to the sorting device 11.

[0016] The sorting device 11 performs sorting based on sorting information notified by the recovery support device 10. For example, the sorting device 11 has a sorting nozzle N, a control function for controlling the sorting nozzle N, and a communication function for communicating with the recovery support device 10. The sorting nozzle N is, for example, a robotic arm having a gripping mechanism at its tip. The sorting nozzle N is driven according to control by a control function and grips the objects to be sorted from the objects to be recovered T being transported on the belt conveyor BC, carries the objects to be recovered T to the top of the recovery container, and puts the objects to be recovered T into the recovery container. The sorting nozzle has, for example, a suction mechanism and grips the objects to be recovered T by sucking them. The recovery container is a container for recovering the objects to be sorted and is, for example, provided within a range that can be transported by the sorting nozzle N near the belt conveyor BC. Furthermore, if the belt conveyor BC branches into multiple lines, the sorting nozzle N may be driven to classify the objects to be recovered T into one of the multiple lines. The control function controls the drive of the sorting nozzle N based on the sorting information. The communication function receives sorting information notified from the recovery support device 10 and outputs the received sorting information to the control function. The sorting nozzle may also have a mechanism for sorting the materials to be recovered T by spraying compressed air, such as an air jet. In this case, the sorting nozzle N can sort the materials to be recovered T by dropping the materials to be sorted into a recovery container located in one area (for example, in front of the belt conveyor BC) and throwing the materials to be recovered T that are not to be sorted into a recovery container located in another area (for example, behind the belt conveyor BC).

[0017] In the collection system 1, the manufacturer server 20 provides an information service to companies (manufacturers) that produce the products to be recycled, providing them with information obtained through collection. The collection system 1 notifies registered manufacturers (registered manufacturers) of aggregated information, such as the amount of collected materials T related to the registered manufacturers' products collected at the sorting plant WS. The recovery support device 10 generates aggregated information. Aggregated information is information obtained by aggregating the information obtained in the process of generating sorting information, for example, information obtained by aggregating the results of sorting the objects to be recovered T. In this embodiment, the recovery system 1 generates aggregated information that is useful to the companies (manufacturers) that manufactured the products corresponding to the objects to be recovered T. For example, the recovery support device 10 aggregates the amount recovered for each manufacturer. Specifically, the recovery support device 10 determines the manufacturers that manufactured the products corresponding to the objects to be recovered T based on the target image G, and generates sorting information that includes information associating the determined manufacturers with the objects to be recovered T. The collection support device 10 transmits the aggregated information to the manufacturer server 20.

[0018] The manufacturer server 20 is a computer managed by the registered manufacturer. The manufacturer server 20 can be a PC, server, cloud, tablet, or the like. The manufacturer server 20 receives aggregated information from the collection support device 10.

[0019] Figure 2 is a block diagram showing the configuration of the collection support device 10. The collection support device 10 includes, for example, a communication unit 101, an acquisition unit 102, a determination unit 103, an aggregation unit 104, an output unit 105, a sorting information storage unit 106, an aggregation information storage unit 107, and a sorting control unit 108. The communication unit 101 communicates with the manufacturer server 20 via the communication network NW. For example, the communication unit 101 sends aggregated information to the manufacturer server 20. Furthermore, the communication unit 101 communicates with the camera C and the sorting device 11. For example, the communication unit 101 receives image information of the target image G from the camera C and outputs the received image information to the acquisition unit 102. The communication unit 101 also transmits sorting information to the sorting device 11. The acquisition unit 102 acquires various types of information. For example, the acquisition unit 102 acquires image information received by the communication unit 101 via the communication unit 101. The acquisition unit 102 outputs the acquired image information to the determination unit 103.

[0020] The determination unit 103 determines the manufacturer of the product corresponding to the recalled item T based on the target image G. For example, the determination unit 103 extracts identification information from the target image G that identifies the product corresponding to the recalled item T, and determines the manufacturer of the product corresponding to the recalled item T based on the extracted identification information.

[0021] The identification information here is information that identifies the product corresponding to the recalled item T, and is represented, for example, by a barcode. The identification information can be any information that can identify the product corresponding to the recalled item T, and may be, for example, a JAN (Japanese Article Number) code. The JAN code has product information embedded in it, and this product information can be used to identify the product. The determination unit 103 can determine the manufacturer that produced the product corresponding to the recalled item T using the product information embedded in the JAN code. Alternatively, the identification information may be printed on the product's packaging or directly attached to the product, such as a product logo, product name, character, or mark. Products often have packaging designs that make them instantly recognizable, and product logos are often placed in prominent locations. If such a product logo is captured in the target image G, the manufacturer of the product corresponding to the recalled item T can be determined based on that product logo.

[0022] For example, the determination unit 103 applies image processing technology to the target image G to perform object recognition and extracts the object to be recovered T as an object captured in the target image G. The determination unit 103 extracts the identification information of each of the object to be recovered T recognized in the target image G from the target image G and determines the manufacturer that produced the product corresponding to the object to be recovered T based on the extracted identification information.

[0023] Figure 3 (Figures 3A and 3B) shows an example of the target image G. Figure 3A shows a target image G1 in which the recalled item T3 is captured. In target image G1, the back of a snack food bag is captured as the recalled item T3. When the determination unit 103 acquires such a target image G1, it extracts, for example, the barcode shown in area E1 of the target image G, and acquires the product information embedded in the extracted barcode using the barcode reading function. Based on the product information, the determination unit 103 determines the manufacturer that produced the product corresponding to the recalled item T. Figure 3B shows a target image G2 in which the object to be collected T4 is captured. In the target image G2, the surface of a snack food bag is captured as the object to be collected T4. When the determination unit 103 acquires such a target image G2, it extracts, for example, the product logo shown in area E2 of the target image G and determines the product corresponding to the extracted product logo.

[0024] The determination unit 103 determines the product corresponding to the product logo using, for example, table information. The table information is information that associates the appearance image of the recalled item T with product attributes (product name, manufacturer name, etc.). For example, the determination unit 103 searches the table information to extract the appearance image that is most similar to the product logo extracted from the target image G2, and obtains the product attributes associated with the extracted appearance image. Based on the product attributes, the determination unit 103 determines the manufacturer that produced the product corresponding to the recalled item T. Alternatively, the determination unit 103 may use machine learning techniques to determine the product corresponding to the product logo. For example, the determination unit 103 inputs the product logo extracted from the target image G2 into a trained model. The trained model here is a model that has learned the correspondence between the appearance of the object to be collected T and its product attributes by training it with training data that associates training images showing the appearance of the object to be collected T with the product attributes of the object to be collected T captured in the training images. By learning this correspondence, the internal parameters of the model are adjusted so that the product attributes of the object to be collected T can be accurately estimated from the appearance image of the object to be collected T. In response to inputting the product logo extracted from the target image G2 into the trained model, the determination unit 103 obtains the estimation result output from the trained model as the product attributes of the object to be collected T. Furthermore, the training images showing the appearance of the object T to be recalled may include not only images of the entire object T, but also images showing only a part of it, for example, images where the packaging material is rolled up or folded, and the product logo is only partially visible. Training images may also include images where a part of the object T, such as the packaging material, has been cut off. By training the model with images showing various possible appearances of the object T during recall, the accuracy of predicting product attributes from appearance images can be improved.

[0025] The determination unit 103 determines the material of the recalled item T based on the determination result of determining the manufacturer of the product corresponding to the recalled item T. For example, manufacturers that produce products often disclose the materials used in the product's packaging on their websites. The determination unit 103 determines the material of the recalled item T by using the manufacturer of the product corresponding to the recalled item T and the information such as the materials used in the product's packaging disclosed by that manufacturer.

[0026] Generally, recyclable plastic products are marked with a recycling identification mark based on the Container and Packaging Recycling Law. These recycling identification marks include, for example, the "PET" mark indicating that it is a recyclable plastic product such as a PET bottle, or the "PLA" mark indicating that it is a recyclable plastic container or packaging other than a PET bottle. Furthermore, below these recycling identification marks, a material identification mark indicating the material may be displayed. Examples of material identification marks include "PP (polypropylene)" and "PE (polyethylene)." For example, the item to be collected, T, may be a plastic product consisting of a bottle body, cap, stopper, and label, used as a container for liquids such as detergents and bleaches. This product's container may have a recycling identification mark, for example, marked "Plastic," and below this recycling identification mark, material identification marks such as "Bottle: PE," "Cap: PE," "Label: PP," and "Stopper: PE" may be shown. This indicates that the material of the bottle body, cap, and stopper is polyethylene (PE), and the material of the label is polypropylene (PP). It is conceivable to use such recycling identification marks and material identification marks to select the items to be collected T. However, these recycling identification marks are often displayed smaller than product logos and are therefore more difficult to recognize. Furthermore, recycling identification marks are not always captured in the target image G. If recycling identification marks are not captured in the target image G, it is difficult to determine the material of the items to be collected T using the recycling identification marks as a clue. In contrast, the collection system 1 of this embodiment identifies the product and manufacturer based on identification information such as barcodes and product logos that can be recognized from the appearance of the object to be collected T captured in the target image G. This makes it possible to accurately determine the material of the object to be collected T even when the recycling identification mark is not captured, or when it is captured but detailed information is difficult to read from the image. Furthermore, some plastic products use multiple materials, and in such cases, only the main material, not all materials, may be indicated by the material identification mark. In such cases, it is difficult to accurately sort plastic products using multiple materials based on the material identification mark. In contrast, the recovery system 1 of this embodiment can identify the product and manufacturer based on the identification information identified from the target image G, making it possible to determine all the materials used in the recovered object T.

[0027] Furthermore, the determination unit 103 may also determine the recycling level of the recovered material T. Here, the recycling level is information indicating the classification of what kind of recycling will be performed on the recovered material T. The recycling level is determined based on the degree of contamination at the time of recovery, etc. Examples of recycling methods include thermal recycling, material recycling, and chemical recycling. Thermal recycling is a recycling method in which the recovered material T is incinerated and the heat (thermal) generated during incineration is reused. Material recycling is a recycling method in which the recovered material T is chemically reduced and substances such as oil, ammonia, and hydrogen are reused. Chemical recycling is a recycling method in which the recovered material T is separated into materials according to its composition and reused as raw materials for regenerating products. For example, PET bottles are classified into categories such as those with little to no dirt, those with slight dirt, and those that are dirty. In the case of PET bottles, the presence or absence of caps and labels, the presence or absence of leftover liquid, and the presence of foreign matter may also be reflected in the above classifications. In the case of plastic products other than PET bottles, they are classified into categories such as those made of a single material and free of dirt, those made of multiple materials and free of dirt, and those made of one or multiple materials and dirty. Those made of a single material and free of dirt are subjected to material recycling or chemical recycling, for example. Those made of multiple materials and free of dirt are subjected to material recycling or chemical recycling, for example. Those made of one or multiple materials and dirty are subjected to thermal recycling. The determination unit 103 determines the degree of contamination of the object to be collected T based on the target image G, and determines the recycling level of the object to be collected T according to the determined degree of contamination. The determination unit 103 also determines the degree of contamination of the object to be collected T based on the color of the object to be collected T. For example, if the difference between the average value of the pixel values ​​in the area where the object to be collected T is imaged in the target image G and the pixel value of a specific color corresponding to the contamination (for example, black) is less than a threshold, it can be determined that the object to be collected T is close to the specific color and is contaminated. Alternatively, the determination unit 103 may determine the degree of contamination of the object to be recalled T based on the difference between the appearance of the product at the time of sale and its appearance at the time of recall. The determination unit 103 can obtain an image showing the appearance of the product at the time of sale from the manufacturer's website or the like. The determination unit 103 can obtain an image showing the appearance at the time of recall from the target image G. The determination unit 103 calculates the difference between the pixel values ​​of the image showing the appearance at the time of sale and the pixel values ​​of the image showing the appearance at the time of recall, and determines that the object to be recalled T is contaminated if the calculated difference is greater than or equal to a threshold. Furthermore, the determination unit 103 may use machine learning techniques to determine the degree of contamination of the object to be collected T, and determine the recycling level of the object to be collected T according to the determined degree of contamination. For example, the determination unit 103 inputs a target image G into a trained model and uses the estimation result obtained as the degree of contamination of the object to be collected T as captured in the target image G. The trained model here is a model that has learned the correspondence between the appearance of the object to be collected T and the degree of contamination of the object to be collected T by training data that associates training images showing the appearance of the object to be collected T with the degree of contamination of the object to be collected T captured in the training images. By learning such a correspondence, the internal parameters of the model are adjusted so that the degree of contamination of the object to be collected T can be accurately estimated from the appearance image of the object to be collected T.

[0028] The determination unit 103 generates sorting information that associates the determination result with the object to be collected T, and stores the generated sorting information in the sorting information storage unit 106. The determination result here includes at least information indicating the manufacturer of the product corresponding to the object to be collected T. The determination result may also include the name of the product corresponding to the object to be collected T, the material used in the object to be collected T, the recycling level, etc. The sorting information may also include information indicating the date and time the object to be collected T was collected, the date and time sorting was performed, the target image G, etc.

[0029] The sorting information storage unit 106 stores sorting information. Figure 4 shows an example of information stored in the sorting information storage unit 106. As shown in this example, the sorting information storage unit 106 stores information corresponding to each of the following items: collection date, target image, product, manufacturer, registration flag, material, and recycling level. The collection date stores information indicating the date on which the item to be collected T was collected. The target image stores image information of the target image G, which is an image of the item to be collected T. The product field stores information indicating the product corresponding to the object to be collected T, as determined by the determination unit 103 based on the target image G. The manufacturer field stores information indicating the manufacturer of the product corresponding to the object to be collected T, as determined by the determination unit 103. The registration flag stores information indicating whether the manufacturer of the product corresponding to the object to be collected T is a registered manufacturer. Here, setting the registration flag to "1" indicates that it is a registered manufacturer, and setting the registration flag to "0 (zero)" indicates that it is an unregistered manufacturer. The material field stores information indicating the material of the object to be collected T, as determined by the determination unit 103. The recycling level field stores information indicating the recycling level of the object to be collected T, as determined by the determination unit 103.

[0030] Returning to the explanation of Figure 2, the sorting control unit 108 controls the sorting device 11 based on the sorting information so that the sorted recovered material T is put into a specific recovery container (a recovery container for recovering the sorted recovered material T).

[0031] For example, consider a sorting plant WS where collected materials T from registered manufacturers are sorted by manufacturer, material, and recycling level, while collected materials T from unregistered manufacturers are sorted by material and recycling level. In this case, the sorting plant WS may have a belt conveyor BC that branches into a line for collecting collected materials T from registered manufacturers and a line for collecting collected materials T from unregistered manufacturers, with collection containers for each material and recycling level installed on each line. When a registered manufacturer recycles materials T collected from its own products at its own recycling plant, the materials T collected by the registered manufacturer's line at the sorting plant WS are transported to the registered manufacturer's recycling plant. On the other hand, materials T collected by a non-registered manufacturer's line at the sorting plant WS are transported to a general-purpose recycling plant that recycles materials according to their respective recycling levels.

[0032] The sorting control unit 108 first controls the sorting device 11 so that each of the items to be recovered T that have been transported on the belt conveyor BC is classified into a line corresponding to either a registered manufacturer or an unregistered manufacturer. The sorting information storage unit 106 reads information indicating the manufacturer of the items to be recovered T and, based on the read information, determines whether the manufacturer of the product corresponding to the item to be recovered T is a registered manufacturer or not. The sorting control unit 108 controls the sorting device 11 to classify the items to be recovered T corresponding to products of registered manufacturers into a line for collecting items to be recovered from registered manufacturers. It also classifies the items to be recovered T corresponding to products of unregistered manufacturers into a line for collecting items to be recovered from unregistered manufacturers. The sorting control unit 108 then controls the sorting device 11 so that each of the materials T classified into each line is sorted according to its material and recycling level. The sorting information storage unit 106 reads information indicating the material and recycling level of the materials T, and controls the sorting device 11 based on the read information to sort the materials T so that they are put into the collection container corresponding to the combination of material and recycling level that corresponds to the materials T.

[0033] The aggregation unit 104 generates aggregation information. For example, the aggregation unit 104 aggregates the amount of collected materials for each registered manufacturer. For example, the aggregation unit 104 measures the weight of each collection container (a collection container that collects the object T corresponding to the product of a registered manufacturer) before and after collection, and calculates the difference (the difference obtained by subtracting the weight before collection from the weight after collection) as the amount of collected materials. Here, the aggregation unit 104 may measure the weight of the collection containers, or the aggregation unit 104 may calculate the amount of collected materials using the weight of the collection containers measured by an external device (for example, a measuring device installed in the sorting plant WS). In this case, the collection support device 10, for example, notifies the external device to measure the weight of the collection containers before and after collection, and in response to this request, the external device notifies the collection support device 10 of the weight of the collection containers. The aggregation unit 104 of the collection support device 10 calculates the amount of collected materials using the weight notified by the external device. The aggregation unit 104 stores the calculated recovery amount as aggregation information in the aggregation information storage unit 107.

[0034] The aggregated information storage unit 107 stores the aggregated information. Figure 5 shows an example of information stored in the aggregated information storage unit 107. As shown in this example, the aggregated information storage unit 107 stores information corresponding to each of the following items: collection period, manufacturer, material, recycling level, and weight. The collection period stores information indicating the period during which the collected item T was collected. A period is set here according to the frequency of notifying the amount of collected items. The frequency of notifying the amount of collected items is determined, for example, according to an agreement with the registered manufacturer. The manufacturer stores information indicating the registered manufacturer. The material stores information indicating the material of the collected item T corresponding to the registered manufacturer's product. The recycling level stores information indicating the recycling level of the collected item T corresponding to the registered manufacturer's product. The weight stores information indicating the amount of collected item T corresponding to the registered manufacturer's product. This figure shows that, for the PP (polypropylene) material of products from registered manufacturer A, W1 [t] of recycled material T with a recycling level of Category 1 was recovered, W2 [t] of recycled material T with a recycling level of Category 2 was recovered, and W3 [t] of recycled material T with a recycling level of Category 3 was recovered.

[0035] In the above explanation, we have used as an example the case in which the amount of recovered material T is calculated by measuring its weight in the collection container, but the explanation is not limited to this. The aggregation unit 104 may also calculate the amount of recovered material based on sorting information. Here, the item to be collected T is collected after the product has been consumed. When the product is consumed, it is expected that the packaging materials, etc., which constitute the item to be collected T, will be removed, separated into multiple pieces, or partially cut off. Therefore, it is likely that the collected item to be collected T will not necessarily be the entirety of the packaging materials used with the product, but rather only a portion of it. When the aggregation unit 104 estimates the weight of the item to be collected T, it is preferable to determine whether the collected item to be collected T is the entirety or a portion of the packaging materials used with the product, and to estimate the weight according to the result of that determination.

[0036] The aggregation unit 104 estimates the weight of the object to be recalled T based on the target image G, for example. For example, the aggregation unit 104 first obtains an overall image of the object to be recalled T and the total weight of the object to be recalled T from product information corresponding to the object to be recalled T disclosed on the manufacturer's website, etc. Next, the aggregation unit 104 compares the overall image of the object to be recalled T obtained from the website, etc. with the object to be recalled T captured in the target image G to determine whether the recovered object to be recalled T is the entirety or only a part of the object to be recalled T used in the product. If the recovered object T is a part of the object T used in the product, the aggregation unit 104 estimates the proportion of the recovered object T that is part of the whole. For example, the aggregation unit 104 estimates the proportion of the recovered object T that is part of the whole by associating feature points extracted from the overall image with feature points extracted from the object T captured in the target image G. Areas corresponding to product logos etc. shown on the packaging material, etc., which are the object T can be extracted as feature points. The aggregation unit 104 classifies the feature points extracted from the overall image into feature points extracted from the recovered object T captured in the target image G and feature points not extracted from the recovered object T captured in the target image G. The aggregation unit 104 uses the ratio of the number of feature points extracted from the recovered object T captured in the target image G to the number of feature points extracted from the overall image as an estimate of the proportion of the recovered object T that is represented in the whole. The aggregation unit 104 multiplies the total weight of the items to be collected T obtained from the website, etc., by an estimated percentage to obtain a multiplied value, which is used as the estimated weight of the collected items T.

[0037] The aggregation unit 104 may also aggregate the contamination rate. The contamination rate is the degree to which two different materials that should be separated and collected separately are mixed together when recycled products are collected. For example, there are two types of pouches used for containers and packaging: plastic pouches made from plastic material, and paper pouches made from paper material, with some or all of the pouch being made from paper instead of plastic. Plastic pouches and paper pouches can sometimes look almost identical, and it is possible to mistakenly put out a paper pouch on the day when plastic resources are collected. If materials other than those intended for collection are mixed in with the materials to be collected, it becomes troublesome during recycling, as the non-collectible materials must be removed. As a countermeasure, the aggregation unit 104 is now configured to aggregate the percentage of collected materials T in relation to the products of registered manufacturers. If a registered manufacturer discovers that plastic pouches and paper pouches are being mixed together in the collection of recycled materials related to their products, they can take measures such as changing the package design or conducting publicity regarding sorting.

[0038] Here, the aggregation unit 104 may also be equipped with an analysis function. In this case, the aggregation unit 104 uses multiple aggregation results to generate information useful to the manufacturer, such as industry trends. For example, the aggregation unit 104 (its analysis function) generates analysis results that show trends in recall volume, relationships with competitors (and their recall volumes), for example, the extraction of similar or different trends, and recall volume for each industry.

[0039] The output unit 105 outputs various information. For example, the output unit 105 outputs the aggregation results from the aggregation unit 104. The output unit 105 also outputs the results of analysis using the aggregation results, based on the analysis function of the aggregation unit 104. For example, when the output unit 105 outputs the amount of recovered data calculated by the aggregation unit 104, the amount of recovered data is notified to the manufacturer server 20 of the corresponding registered manufacturer.

[0040] The storage unit (including the sorting information storage unit 106 and the aggregation information storage unit 107) of the recovery support device 10 is composed of a storage medium, such as an HDD (Hard Disk Drive), flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), RAM (Random Access read / write Memory), ROM (Read Only Memory), or any combination of these storage media. This storage unit can use, for example, non-volatile memory.

[0041] Here, we will explain the processes performed by the recovery system 1 using Figures 6 and 7. Figure 6 is a sequence diagram showing the flow of the recovery process performed by the recovery system 1. Figure 7 shows the flow of the sorting process shown in step S14 of Figure 6.

[0042] The entire recovery process will be explained using Figure 6. The recovery company 30 of the recovery system 1 recovers the materials T to be recovered (step S10). The materials T recovered by the recovery company 30 are fed onto the belt conveyor BC of the sorting line at the sorting plant WS (step S11). The recovery support device 10 images the materials T being transported on the belt conveyor BC (step S12). Based on the target image G, the recovery support device 10 generates sorting information by determining the manufacturer of the product corresponding to the materials T, the material of the materials T, the recycling level of the materials T, etc. (step S13). Based on the sorting information, the recovery support device 10 controls the sorting device 11 and performs sorting (step S14). At the sorting plant WS, processing is carried out for each sorted material T (step S15). For example, materials T recovered by material and recycling level at a registered manufacturer's line are transported to the registered manufacturer's recycling plant. The collected materials T, sorted by material and recycling level at the production lines of unregistered manufacturers, are transported to general-purpose recycling plants where they are recycled according to their respective material and recycling levels.

[0043] Using Figure 7, the sorting process shown in step S14 of Figure 6 will be explained. The sorting control unit 108 of the recovery support device 10 determines whether or not the sorting information of the object to be recovered T indicates that the manufacturer has been determined (step S140). If it indicates that the manufacturer has been determined, the sorting control unit 108 determines whether or not the manufacturer in the sorting information is a registered manufacturer (step S141), and if it is a registered manufacturer, it controls the sorting device 11 so that the object to be recovered T is classified into the line of the registered manufacturer (step S142). On the other hand, if the manufacturer of the product corresponding to the object to be recovered T has not been determined, or if the manufacturer of the product corresponding to the object to be recovered T has been determined but is not a registered manufacturer, the sorting control unit 108 controls the sorting device 11 so that the object to be recovered T is classified into the line of an unregistered manufacturer (step S143). Next, the sorting control unit 108 determines whether the sorting information of the object to be recovered T indicates that the material has been determined (step S144). If the material has been determined, the sorting control unit 108 determines whether the sorting information of the object to be recovered T indicates that the recycling level has been determined (step S145). If the recycling level has been determined, the sorting control unit 108 controls the sorting device 11 so that the object to be recovered T is put into collection containers according to material and recycling level (step S146). On the other hand, if step S145 does not indicate that the recycling level has been determined in the sorting information of the object to be recovered T, the sorting control unit 108 controls the sorting device 11 so that the object to be recovered T is put into a collection container categorized by material (without determining the recycling level) (step S147). Also, if step S144 does not indicate that the material has been determined in the sorting information of the object to be recovered T, the sorting control unit 108 controls the sorting device 11 so that the object to be recovered T is put into a collection container for unknown material (step S147).

[0044] The above flowchart illustrates a case where the sorting line using a belt conveyor BC branches into lines for registered manufacturers and lines for unregistered manufacturers, but it is not limited to this. Separate collection containers for registered and unregistered manufacturers, each categorized by material and recycling level, may be provided on a single common sorting line. Furthermore, whether or not to determine the material and whether or not to determine the recycling level may be arbitrarily set depending on the acceptance method of the collected material T at the selected transport destination recycling plant.

[0045] Furthermore, the above describes an example in which the manufacturer of the product corresponding to the recalled item T is identified and the amount of recalled items for each identified manufacturer is aggregated. However, the system is not limited to this. In addition to the amount of recalled items per manufacturer, the system may also aggregate the amount of recalled items for each further subdivided product, the amount of recalled items for each material, and the amount of recalled items for each type of product corresponding to the recalled item T (food, toiletries, etc.). In this case, the determination unit 103 is configured to determine which category the recalled item T belongs to, based on any category to which the recalled item T can be classified, such as the product corresponding to the recalled item T, the type of product, the material of the recalled item T, and combinations thereof. That is, the determination unit 103 determines the category corresponding to the recalled item T based on the target image G. The aggregation unit 104 aggregates the amount of recalled items for each category determined by the determination unit 103. As a result, the recall support device 10 of the embodiment can provide information that classifies and aggregates the recalled items T into any category.

[0046] As described above, according to the first embodiment, the collection support device 10 is installed in a sorting plant WS (an example of a collection base for a recycling company that recycles the collected items T, including packaging materials). The collection support device 10 comprises an acquisition unit 102, a determination unit 103, an aggregation unit 104, and an output unit 105. The acquisition unit 102 acquires a target image G. The target image G is an image of the collected items T that have been collected at a collection base and transported to the sorting plant WS by a recycling company. The determination unit 103 determines the category corresponding to the collected items T based on the target image G, for example, the manufacturer of the product corresponding to the collected items T. The aggregation unit 104 aggregates the amount of collected items for each category. Thus, the collection support device 10 of the first embodiment can determine the category of the collected items T based on the target image G, for example, the manufacturer of the product corresponding to the collected items T. For example, it is possible to aggregate the amount of collected items from each manufacturer from among a mixture of collected items T originating from products of various manufacturers and notify manufacturers that have set recycling targets. Furthermore, since it includes materials collected at collection points established by local governments and other organizations and consolidated at sorting plants (WS), there is no need to go to dedicated collection boxes set up by manufacturers for their own products. Therefore, it is possible to reduce the burden on users regarding resource collection while utilizing the information obtained from the collection process.

[0047] Furthermore, in the first embodiment described above, the determination unit 103 extracts identification information from the target image G that identifies the product corresponding to the recalled object T, which is identification information attached to the recalled object T. As identification information, one-dimensional codes or two-dimensional codes such as JAN codes with embedded product information can be used. Alternatively, the package design printed on the recalled object, such as a product logo, can also be used as identification information. Based on the extracted identification information, the determination unit 103 determines the manufacturer of the product corresponding to the recalled object T. Thus, in the recall support device 10 of the first embodiment, the product corresponding to the recalled object T and the manufacturer of that product can be easily and accurately identified using product information embedded in barcodes attached to the recalled object T, or a conspicuously designed product logo, as clues.

[0048] Furthermore, in the first embodiment described above, the determination unit 103 determines the manufacturer of the product corresponding to the object to be recovered by inputting the target image G into a learning model that has learned the correspondence between a learning image in which the object to be recovered T is captured in the learning image and information indicating the manufacturer of the product corresponding to the object to be recovered T captured in the learning image. This makes it possible to easily identify the product corresponding to the object to be recovered T and the manufacturer of that product from the target image G using the learned model.

[0049] Furthermore, in the first embodiment described above, the acquisition unit 102 acquires a target image G obtained by imaging the object to be recovered T being transported by the belt conveyor BC of the sorting plant WS from above. As a result, the recovery support device 10 of the first embodiment can image the object to be recovered T using the sorting line used in the sorting plant WS, and there is no need to provide dedicated equipment for obtaining the target image G.

[0050] Furthermore, in the first embodiment described above, the determination unit 103 determines the material of the object to be recovered T from the product corresponding to the object to be recovered T determined based on the target image G. The aggregation unit 104 aggregates the amount recovered for each material. As a result, the recovery support device 10 of the first embodiment can notify the manufacturer of which materials and in what quantities have been recovered from the recyclable materials used in the manufacturer's products, and can provide the manufacturer with meaningful information.

[0051] Furthermore, in the first embodiment described above, the determination unit 103 determines the material of the object T to be recovered, and the aggregation unit 104 aggregates the mixing rate, which is the ratio of the amount of paper material recovered to the amount of plastic material recovered. As a result, the recovery support device 10 of the first embodiment can aggregate the degree to which different materials, such as plastic material and paper material, are mixed together during recovery. By notifying registered manufacturers of this degree (purchase rate), meaningful information can be provided to the manufacturers.

[0052] Furthermore, in the first embodiment described above, the determination unit 103 determines the recycling level of the object to be collected T based on the color information of the target image G. The aggregation unit 104 aggregates the amount of collected material for each recycling level. As a result, the collection support device 10 of the first embodiment can aggregate which method of recycling the collected object T will be recycled by, among thermal recycling, material recycling, chemical recycling, etc. By notifying registered manufacturers of the recycling level, meaningful information can be provided to the manufacturers.

[0053] (Second embodiment) A second embodiment of the present invention will be described below. In the second embodiment, information is provided that takes into account the marketing of the registered manufacturer. Specifically, aggregated information is generated that associates the region where the recalled item T was collected with the amount of recalled item T related to the registered manufacturer's product.

[0054] While manufacturers produce products, it is generally the retailers, not the manufacturers themselves, that sell them. Therefore, unless it's a special store like a manufacturer's own retail outlet that can share POS (Point of Sale) information, it's difficult for manufacturers to accurately track how much of their products are being sold in stores. Furthermore, even if POS information from stores is available, it's difficult to determine where customers who purchased the product are consuming it. For example, it's difficult to determine where a popular souvenir confectionary from region A is being consumed. If a manufacturer knows which regions consume the most of their products, they can develop marketing strategies such as considering product specifications tailored to those regions. For example, if a confectionery product exclusive to region A is found to be most consumed in region B, which is far from region A, the manufacturer can develop strategies that take consumption in region B into consideration.

[0055] As a countermeasure, the recovery support device 10 of this embodiment generates aggregated information that associates the area where the object T to be recovered was recovered with the amount recovered. Specifically, the acquisition unit 102 acquires location information indicating the location of the collection point when collecting the materials T to be collected at the sorting plant WS. The location information may also be information indicating the region, city, town, or village of the collection point. For example, when the materials T to be collected are transported to the sorting plant WS, the location information of the collection point is notified to the collection support device 10 via a work terminal owned by a worker in the collection vehicle V. The acquisition unit 102 acquires the location information notified to the collection support device 10 from the work terminal. The determination unit 103 generates sorting information including location information and stores the generated sorting information in the sorting information storage unit 106. Figure 8 shows another example of information stored in the sorting information storage unit 106. In the second embodiment, as shown in this figure, location information corresponding to the regional item is stored in the sorting information storage unit 106. The regional item stores information indicating the location of the collection point where the object to be collected T was collected. The aggregation unit 104 generates aggregated information including location information and stores the generated information in the aggregated information storage unit 107. Figure 9 shows another example of information stored in the aggregated information storage unit 107. In the second embodiment, as shown in this figure, the aggregated information storage unit 107 stores information indicating the amount of recalled material T used in the registered manufacturer's products, broken down by region.

[0056] As described above, in the second embodiment of the recovery support device 10, the acquisition unit 102 acquires location information indicating the location of the recovery base where the object to be recovered T was recovered. The aggregation unit 104 aggregates the amount recovered for each region based on the location information. As a result, the recovery support device 10 of the second embodiment can generate aggregate information that associates the region where the object to be recovered T was recovered with the amount recovered, and can provide manufacturers with meaningful information, such as notifying registered manufacturers of information showing how much of their products are being consumed in which region.

[0057] (Third embodiment) A third embodiment of the present invention will be described below. In the third embodiment, information is provided that takes into account the marketing of the registered manufacturer. Specifically, aggregated information is generated showing the time-series changes in the amount of recalled material T related to the registered manufacturer's product. Generally, it's clear that sales of a campaigned product increase when a campaign is implemented. However, even with POS data, it's impossible to determine how long it takes for those products to be consumed. For example, it's difficult to know whether a product whose sales increased through a spring campaign was consumed immediately, or whether it was stored in the customer's home and consumed after a period of about six months. If manufacturers know whether their products purchased for home use are consumed immediately or after being stored for a certain period, they can develop marketing strategies, such as considering the timing of campaigns, based on that timeframe. As a countermeasure, the collection support device 10 of this embodiment generates aggregated information showing the time-series changes in the amount of collected items T corresponding to the campaign target products. Specifically, the determination unit 103 determines the product corresponding to the recalled item T, and also determines whether this recalled item T corresponds to a campaign target product. For example, in a campaign, product packaging is often changed for the campaign, or a campaign-issued JAN code is assigned. Based on such changes, the determination unit 103 determines whether the recalled item T corresponds to a campaign target product. For example, the determination unit 103 can determine whether the product corresponding to the recalled item T is a campaign target product by using product information embedded in the JAN code extracted from the target image G. Alternatively, the determination unit 103 can determine whether the product corresponding to the recalled item T is a campaign target product based on the package design extracted from the target image G. The aggregation unit 104 generates aggregated information showing the time-series changes in the amount of collected items T that have been determined to be campaign-eligible products, and stores the generated information in the aggregated information storage unit 107. Figure 10 shows another example of information stored in the aggregated information storage unit 107. In Figure 10, the horizontal axis shows the recovery period and the vertical axis shows the recovery amount. In the third embodiment, as shown in this figure, the aggregated information storage unit 107 stores information showing the time-series change in the amount of recovered material T used in the registered manufacturer's products. In addition, in collection system 1, the time-series changes in the amount of collected materials may be shown for each collection point as shown in Figure 10, or the time-series changes in the amount of collected materials may be shown aggregated by region. Furthermore, the system may be configured to allow registered manufacturers to set whether to display aggregated information showing the time-series changes in the collected amount for all collection points combined. In this case, the determination unit 103 generates sorting information that associates collection points with regions using a method similar to that of the second embodiment. The aggregation unit 104 generates aggregated information showing the collected amount in time series for each collection point where the object to be collected T was collected. The aggregation unit 104 generates aggregated information for display, which includes an input field for setting location information to indicate a region, on a graph showing the time-series changes in the collected amount as shown in Figure 10. In this aggregated information for display, the collected amount from collection points in the region corresponding to the set location information is aggregated according to the location information set in the input field, and a graph showing the time-series changes in the aggregated collected amount is displayed. This allows registered manufacturers to grasp the time-series changes in the collected amount in regions corresponding to the regions where campaigns were conducted.

[0058] As described above, in the third embodiment of the collection support device 10, the determination unit 103 determines from the target image G whether the product corresponding to the item to be collected T is a campaign target product. The aggregation unit 104 aggregates the time-series changes in the amount of collected items T corresponding to the campaign target products. As a result, the collection support device 10 of the third embodiment can generate aggregated information that correlates whether products purchased by users tend to be consumed immediately, or the areas where the items to be collected T are collected with the amount collected, thereby providing manufacturers with meaningful information.

[0059] The functions of the recovery support device 10 in the above-described embodiment may be implemented using a computer. In that case, the functions may be implemented by recording a program for implementing these functions on a computer-readable recording medium, loading the program recorded on this recording medium into a computer system, and executing it. Here, "computer system" includes hardware such as an OS and peripheral devices. Furthermore, "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, CD-ROMs, and storage devices such as hard disks built into a computer system. Moreover, "computer-readable recording medium" may also include those that dynamically hold programs for a short period of time, such as communication lines used when transmitting programs via networks such as the Internet or communication lines such as telephone lines, and those that hold programs for a certain period of time, such as volatile memory inside a computer system that acts as a server or client in such cases. Furthermore, the above-mentioned program may be for implementing a part of the functions described above, or it may be a program that can implement the above-mentioned functions in combination with a program already recorded in the computer system, or it may be implemented using a programmable logic device such as an FPGA (Field Programmable Gate Array).

[0060] Furthermore, the program that operates in the recovery support device 10 in one aspect of the present invention may be one or more programs that control a processor such as a CPU (Central Processing Unit) (a program that makes a computer function) in order to realize the functions shown in the above embodiments and modifications relating to one aspect of the present invention. The term "computer" here includes quantum computers. The information handled by each of these devices is temporarily stored in RAM (Random Access Memory) during processing, and then stored in various storage devices such as flash memory and HDD (Hard Disk Drive), and may be read, modified, and written by the CPU or the like as needed.

[0061] Furthermore, some or all of the recovery support devices 10 in each of the embodiments and modifications described above may be implemented using a computer equipped with one or more processors. In that case, the program for implementing this control function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read by a computer system and executed.

[0062] Furthermore, some or all of the recovery support devices 10 in each of the embodiments and modifications described above may be implemented as LSIs, which are typically integrated circuits, or as a chipset. Also, each functional block of the recovery determination device 100 in each of the embodiments and modifications described above may be individually chipped, or some or all of them may be integrated into a chip. In addition, the method of implementing the integrated circuit is not limited to LSIs, but may also be implemented using dedicated circuits and / or general-purpose processors. Furthermore, if an integrated circuit technology that can replace LSIs emerges due to advances in semiconductor technology, it is also possible to use an integrated circuit based on that technology.

[0063] Although various embodiments and modifications have been described in detail above with reference to the drawings as one aspect of this invention, the specific configuration is not limited to these embodiments and modifications, and includes design changes and the like that do not depart from the gist of this invention. Furthermore, various modifications are possible within the scope shown in the claims for one aspect of this invention, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of this invention. In addition, configurations in which elements described in the above embodiments and modifications that produce similar effects are substituted for each other are also included. [Explanation of Symbols]

[0064] 1...Collection system, 10...Collection support device, 20...Manufacturer server, 30...Collection company, 101...Communication unit, 102...Acquisition unit, 103...Determination unit, 104...Aggregation unit, 105...Output unit, 106...Sorting information storage unit, 107...Aggregation information storage unit, G...Target image, T...Item to be collected, WS...Sorting plant

Claims

1. It is located at a collection point for recycling companies that recycle collected materials, including packaging materials. An acquisition unit that acquires target images of the collected items that have been collected at a collection point and transported to the collection point by a collection company, A determination unit that determines the category corresponding to the object to be collected based on the target image, A summation unit that aggregates the amount of each category determined by the determination unit, Having, Recovery support device.

2. The determination unit determines the manufacturer of the product corresponding to the item to be collected as the category, The aggregation unit aggregates the recall amounts for each manufacturer determined by the determination unit. The recovery support device according to claim 1.

3. The system further includes an output unit that outputs the aggregated results compiled by the aforementioned aggregation unit. The recovery support device according to claim 1.

4. The determination unit extracts identification information from the target image that identifies the product corresponding to the object to be collected, and determines the category based on the extracted identification information. The recovery support device according to claim 1.

5. The determination unit determines the category by using the JAN code, which has product information embedded in it attached to the object to be recalled, as the identification information. The recovery support device according to claim 4.

6. The determination unit determines the category by using the package design printed on the object to be collected as the identification information. The recovery support device according to claim 4.

7. The determination unit determines the manufacturer of the product corresponding to the collected object by inputting the target image into a learning model that has learned the correspondence between a learning image in which the collected object is captured and information indicating the manufacturer of the product corresponding to the collected object captured in the learning image. The recovery support device according to claim 1.

8. The acquisition unit acquires the target image obtained by taking an image of the object to be collected from above as it is transported by the belt conveyor of the aggregation base. The recovery support device according to claim 1.

9. The determination unit determines the recycling level of the object to be collected based on the color information of the target image. The aforementioned aggregation unit aggregates the amount of collected items for each recycling level. The recovery support device according to claim 1.

10. The acquisition unit acquires location information indicating the location of the collection point where the object to be collected was collected. The aggregation unit aggregates the amount collected for each region based on the location information. The recovery support device according to claim 1.

11. The determination unit determines whether the product using the recalled material is a campaign product, The aggregation unit aggregates the time-series changes in the amount of recalled items for products corresponding to the campaign target products. The recovery support device according to claim 1.

12. The recovery support device according to claim 1, A sorting device that sorts according to the categories determined by the determination unit, A recovery system having

13. A collection support method performed by a collection support device, which is a computer installed at a collection center of a recycling company that recycles materials including packaging materials, A target image is obtained of the object to be collected at the collection point and transported to the collection point by the collection company. Based on the aforementioned target image, the category corresponding to the object to be collected is determined. The amount of each category determined above is aggregated. Collection support methods.

14. A program to be executed on a computer installed at a collection point of a recycling company that recycles collected materials, including packaging materials, A target image is obtained of the object to be collected at the collection point and transported to the collection point by the collection company. Based on the aforementioned target image, the category corresponding to the object to be collected is determined. The amount of each category determined above is aggregated. A program that includes the following.

Images