Sorting system, reuse processing apparatus, recycling processing apparatus, disposal processing apparatus, sorting method, and program

Abstract

To provide a sorting system and the like suitable for realizing sorting that focuses on a value of each collected item for collected items such as clothing.SOLUTION: A sorting system 1 includes a sorting apparatus 5, a reuse processing apparatus 11 that evaluates a resale value of a collected item, a recycling processing apparatus 13 that collects fibers from the collected item, and a disposal processing apparatus 15. The sorting apparatus 5 includes a conveying device 31 and a determination processing unit 45. The determination processing unit 45 determines whether or not the collected item is a fiber product by using image data obtained by imaging the collected item. The conveying device 31 conveys the collected item to one of the reuse processing apparatus 11, the recycling processing apparatus 13, and the disposal processing apparatus 15 when the determination processing unit 45 determines that the collected item is a fiber product. When the determination processing unit 45 determines that the collected item is not a fiber product, the conveying device conveys the collected item to one of the recycling processing apparatus 13 and the disposal processing apparatus 15.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a sorting system, a sorting method, and a program, and more particularly to a sorting system for sorting recovered items. [Background technology]

[0002] Patent Document 1 describes a method of guessing and sorting multiple waste materials. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-195751 Summary of the Invention [Problem to be solved by the invention]

[0004] The waste materials described in Patent Document 1 are treated as having no value individually, so it is important that the waste materials are ultimately grouped together according to their materials.

[0005] However, for example, clothing that one person considers worthless because they have outgrown it may be valuable to another. Because such value varies from person to person, workers visually evaluate and sort the items, but it is difficult for workers to comprehensively evaluate specific brands, and there is also a limit to the amount that can be sorted.

[0006] Therefore, an object of the present invention is to provide a sorting system suitable for sorting collected items such as clothing with a focus on the value of each collected item. [Means for solving the problem]

[0007] A first aspect of the present invention is a sorting system for sorting recovered items, comprising a sorting device, a reuse processing device that evaluates the resale value of the recovered items, a recycling processing device that recovers fibers from the recovered items, and a waste processing device, wherein the sorting device comprises a conveying device and a discrimination processing unit, wherein the discrimination processing unit determines whether the recovered item is a textile product using image data obtained by photographing the recovered item, and wherein the conveying device, if the discrimination processing unit determines that the recovered item is a textile product, conveys the recovered item to either the reuse processing device, the recycling processing device, or the waste processing device, and if the discrimination processing unit determines that the recovered item is not a textile product, conveys the recovered item to either the recycling processing device or the waste processing device.

[0008] A second aspect of the present invention is a sorting system of the first aspect, wherein the reuse processing device comprises a proximity measurement device and a value assessment device, the value assessment device comprises a proximity analysis processing unit and a value assessment processing unit, the proximity measurement device measures the recovered items at a distance closer than the distance at which the recovered items were photographed to obtain the image data, the proximity analysis processing unit uses the measurement data of the proximity measurement device to identify parts of the recovered items that have been damaged during use, and the value assessment processing unit assesses the value of the recovered items by reference to the damaged parts identified by the proximity analysis processing unit.

[0009] A third aspect of the present invention is a sorting system according to the first or second aspect, wherein the discrimination processing unit determines whether the collected items are clothing using the image data, the recycling processing unit includes an accessory measuring device and a collection preparation device, and when the discrimination processing unit determines that the collected items are clothing, it obtains data indicating the accessories present in each part of the collected items, the accessory measuring device is capable of measuring the material of the accessories separately from the fabric, and the accessory measuring device uses the data indicating the accessories present in each part of the collected items to perform measurement processing according to the tendency of accessories to be used in each part, and the collection preparation device removes the accessories detected by the measurement by the accessory measuring device from the collected items.

[0010] A fourth aspect of the present invention is a sorting system according to any one of the first to third aspects, wherein the recycling processing device comprises a material measuring device, a recycling analysis device, a collection preparation device, and a fiber recovery device, wherein the material measuring device measures the material of the recovered product, and the recycling analysis device determines, as a first determination, whether the material is close to a plurality of target materials through multiple analyses, and if the analysis result of the material is close to at least one of the target materials, then, as a second determination, estimates the proportion of the target materials, and the collection preparation device cuts the recovered product, and the fiber recovery device performs fiber recovery processing on the cut recovered product according to the detected target materials.

[0011] A fifth aspect of the present invention is a sorting system according to any one of the first to fourth aspects, comprising a pre-processing device, the pre-processing device comprising an unfolding device, a camera, and a sensor, the unfolding device unfolds the collected items, the camera photographs the collected items unfolded by the unfolding device to obtain the image data, the sensor measures the collected items unfolded by the unfolding device, and the discrimination processing unit uses the image data and measurement data obtained by the camera and the sensor to determine whether the collected items are clothing, and if it is determined that the collected items are clothing, obtains data indicating the category into which the clothing is classified, the parts that make up the clothing, and the accessories present for each of the parts.

[0012] A sixth aspect of the present invention is the sorting system of the fifth aspect, wherein the unfolding device comprises a lower conveying device and an upper conveying device, the upper conveying device being positioned above the lower conveying device, and the upper conveying device transports the uppermost of the stacked recovered items in the opposite direction to the direction in which the lower conveying device transports the recovered items.

[0013] A seventh aspect of the present invention is a reuse processing device for evaluating the resale value of recovered items, comprising a proximity measurement device and a value evaluation device, wherein the value evaluation device comprises a proximity analysis processing unit and a value evaluation processing unit, wherein the proximity measurement device measures the recovered item at a distance closer than the distance at which the recovered item was photographed to obtain image data, the proximity analysis processing unit uses the measurement data from the proximity measurement device to identify areas of the recovered item that have been damaged during use, and the value evaluation processing unit evaluates the value of the recovered item by referring to the image data and the damaged areas identified by the proximity analysis processing unit.

[0014] An eighth aspect of the present invention is a recycling processing device that recovers fibers from collected items, comprising a material measuring device, an accessory measuring device, a collection preparation device, and a fiber recovery device, wherein when the collected items are determined to be clothing, data indicating the accessories present for each part of the collected items is generated, the material measuring device measures the material of the collected items, and the accessory measuring device is capable of measuring the material of the accessories separately from the fabric, the accessory measuring device uses the data indicating the accessories present for each part of the collected items to perform measurement processing according to the tendency of accessories to be used for each part, the collection preparation device removes accessories detected by measurement by the accessory measuring device from the collected items, the collection preparation device cuts the collected items, and the fiber recovery device performs fiber recovery processing on the cut collected items according to the detected material to be determined.

[0015] A ninth aspect of the present invention is a waste disposal device that includes a waste disposal device that discards recovered items that a sorting device has determined cannot be reused or recycled, and the sorting device determines whether the recovered items can be reused or recycled, or both, based on image data and measurement data obtained by photographing and measuring the recovered items with a camera and a sensor.

[0016] A tenth aspect of the present invention is a sorting method in a sorting system for sorting recovered goods, the sorting system comprising a sorting device, a reuse processing device that evaluates the resale value of the recovered goods, a recycling processing device that recovers fibers from the recovered goods, and a waste processing device, the sorting device comprising a conveying device and a discrimination processing unit, the discrimination processing unit including a step of determining whether the recovered goods are textile products using image data obtained by photographing the recovered goods, and a step of conveying the recovered goods to one of the reuse processing device, the recycling processing device, or the waste processing device if the discrimination processing unit determines that the recovered goods are textile products, and a step of conveying the recovered goods to one of the recycling processing device or the waste processing device if the discrimination processing unit determines that the recovered goods are not textile products.

[0017] An eleventh aspect of the present invention is the sorting method of the tenth aspect, wherein the reuse processing device comprises a proximity measurement device and a value assessment device, the value assessment device comprises a proximity analysis processing unit and a value assessment processing unit, the proximity measurement device measures the recovered items at a distance closer than the distance at which the recovered items were photographed to obtain the image data, the proximity analysis processing unit uses the measurement data of the proximity measurement device to identify parts of the recovered items that have been damaged during use, and the value assessment processing unit evaluates the value of the recovered items by reference to the damaged parts identified by the proximity analysis processing unit.

[0018] A twelfth aspect of the present invention is the sorting method of the tenth or eleventh aspect, wherein the discrimination processing unit determines whether the collected items are clothing using the image data, the recycling processing unit includes an accessory measuring device and a collection preparation device, and when the discrimination processing unit determines that the collected items are clothing, it obtains data indicating the accessories present for each part of the collected items, the accessory measuring device is capable of measuring the material of the accessories separately from the fabric, the accessory measuring device uses the data indicating the accessories present for each part of the collected items to perform measurement processing according to the tendency of accessories to be used for each part, and the collection preparation device removes the accessories detected by the measurement by the accessory measuring device from the collected items.

[0019] A thirteenth aspect of the present invention is a sorting method according to any one of the tenth to twelfth aspects, wherein the recycling processing device comprises a material measuring device, a recycling analysis device, a collection preparation device, and a fiber recovery device, wherein the material measuring device measures the material of the recovered product, and the recycling analysis device determines, as a first judgment, whether the material is close to a plurality of target materials through multiple analyses, and if the analysis result of the material is close to at least one of the target materials, as a second judgment, estimates the proportion of the target materials, and the collection preparation device cuts the recovered product, and the fiber recovery device performs fiber recovery processing on the cut recovered product according to the detected target materials.

[0020] A fourteenth aspect of the present invention is a sorting method according to any one of the tenth to thirteenth aspects, comprising a pre-processing device, the pre-processing device comprising an unfolding device, a camera, and a sensor, the unfolding device unfolds the collected items, the camera photographs the collected items unfolded by the unfolding device to obtain the image data, the sensor measures the collected items unfolded by the unfolding device, and the discrimination processing unit uses the image data and measurement data obtained by the camera and the sensor to determine whether the collected items are clothing, and if it is determined that the collected items are clothing, obtains data indicating the category into which the clothing is classified, the parts that make up the clothing, and the accessories present for each of the parts.

[0021] A fifteenth aspect of the present invention is a program for causing a computer to function as the discrimination processing unit of any one of the first to sixth aspects. The present invention may also be understood as a computer-readable recording medium on which the program of the fifteenth aspect is recorded. Furthermore, each unit constituting the present invention may be realized by technology such as AI, for example. [Effects of the Invention]

[0022] According to each aspect of the present invention, if a collected item is determined to be clothing, it can be sorted into a reuse processing device to evaluate the resale value of the clothing, and sorting can be realized with a focus on the value of each collected item. On the other hand, if a collected item is determined to be clothing but has no recognized value as clothing, or if the collected item is determined not to be clothing at all, it can be sorted into a recycling processing device to recover fibers from the collected item, or sorted into a waste processing device for disposal. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a block diagram showing an example of a configuration of a sorting system 1 according to an embodiment of the present invention. [Figure 2] FIG. 10 is a flowchart showing an example of a learning process for generating a model used for determination by a discrimination processing unit 45. [Figure 3] 1 is a block diagram for explaining a sorting process of collected items (clothes, etc.) in the sorting system 1. FIG. [Figure 4] 1 is a flow diagram for explaining a sorting process of collected items (clothes, etc.) in the sorting system 1. FIG. [Figure 5] 1A is a block diagram showing an example of the configuration of a reuse processing device 11, and FIG. 1B is a flow diagram showing an example of the operation thereof. [Figure 6] 1A is a block diagram showing an example of the configuration of a recycle processing device 13, and FIG. 1B is a flow diagram showing an example of the operation thereof. [Figure 7]1A is a block diagram showing an example of the configuration of a waste processing device 15, and FIG. 1B is a flow diagram showing an example of the operation thereof. [Figure 8] FIG. 2 is a diagram for explaining an example of a specific configuration of the unfolding device 21, the camera 23, and the sensor 25 in the pre-processing device 3 of FIG. [Figure 9] FIG. 11 is a first diagram for explaining another example of the specific configuration of the unfolding device 21, the camera 23, and the sensor 25 in the pre-processing device 3 of FIG. [Figure 10] FIG. 2 is a second diagram for explaining another example of the specific configuration of the unfolding device 21, the camera 23, and the sensor 25 in the pre-processing device 3 of FIG. [Figure 11] FIG. 1 is a diagram showing a camera (an example of a combination of the camera 23 and the discrimination processing unit 45 in FIG. 1) incorporating image recognition AI (artificial intelligence) developed by the inventor. [Figure 12] 10 is a diagram showing an example of a discrimination result by a discrimination processing unit 45. FIG. [Figure 13] 10 is a diagram showing an example of the result of discrimination of an accessory by a discrimination processing unit 45. FIG. [Figure 14] FIG. 6 is a diagram illustrating an example of a proximity measurement device 51 in FIG. [Figure 15] FIG. 7 is a first diagram showing an example of analysis of materials by the recycling analysis device 67 of FIG. 6. [Figure 16] FIG. 7 is a second diagram showing an example of analysis of materials by the recycling analysis device 67 of FIG. 6. DETAILED DESCRIPTION OF THE INVENTION

[0024] Hereinafter, examples of the present invention will be described with reference to the drawings, but the present invention is not limited to the following examples. [Example]

[0025] FIG. 1 is a block diagram showing an example of the configuration of a sorting system 1 according to an embodiment of the present invention.

[0026] The sorting system 1 includes a pre-processing device 3, a sorting device 5, a reuse processing device 11, a recycle processing device 13, and a waste processing device 15.

[0027] The preprocessing device 3 includes a spreading device 21, a camera 23, a sensor 25, and a preprocessing analysis device 27. The spreading device 21 spreads out an object (such as clothing). The camera 23 and the sensor 25 photograph and measure the object spread out by the spreading device 21, respectively. The spreading device 21 sends the photographed and measured object to the sorting device 5. The preprocessing analysis device 27 transmits the image data and measurement data obtained by the camera 23 and the sensor 25 to the sorting device 5. Note that the preprocessing device 3 may obtain information about the material used in the object using attached information attached to the object. For example, this information may be obtained by using the camera 23 to photograph a code (e.g., a two-dimensional code) printed on a tag or the back of a fabric, or by using the sensor 25 to read a chip-like or thread-like RFID tag attached to the object.

[0028] The sorting device 5 includes a transport device 31 and a sorting control device 33. The sorting control device 33 uses image data and measurement data obtained by the camera 23 and sensor 25 to determine whether the object sent from the pre-processing device 3 should be sent to the reuse processing device 11, the recycle processing device 13, or the waste processing device 15. The transport device 31 transports the object sent from the pre-processing device 3 to one of the reuse processing device 11, the recycle processing device 13, or the waste processing device 15, according to the determination result by the sorting control device 33.

[0029] The reuse processing device 11, the recycle processing device 13, and the waste processing device 15 process the objects sent from the sorting device 5 for reuse, recycling, and waste, respectively.

[0030] The sorting control device 33 includes a sorting storage unit 35 and a sorting processing unit 37. The sorting storage unit 35 is realized by a storage device such as a memory, and stores data necessary for processing by the sorting processing unit 37. The sorting processing unit 37 is realized by a processing device (processor) that operates under the control of a program, and performs processing to determine whether the object sent from the preprocessing device 3 should be sent to the reuse processing device 11, the recycle processing device 13, or the waste processing device 15.

[0031] The selection processing unit 37 includes a preprocessing result acquisition unit 39, a teacher processing unit 41, a learning processing unit 43, and a discrimination processing unit 45. The preprocessing result acquisition unit 39, the teacher processing unit 41, the learning processing unit 43, and the discrimination processing unit 45 are realized by, for example, a processing device (processor) that operates under the control of a program.

[0032] The discrimination processing unit 45 uses AI (artificial intelligence) to determine whether the object sent from the pre-processing unit 3 should be sent to the reuse processing unit 11, the recycling processing unit 13, or the waste processing unit 15. For example, the discrimination processing unit 45 determines whether the object is a textile product using image data obtained by photographing the object. If the discrimination processing unit 45 determines that the object is a textile product, the transport device 31 transports the object to either the reuse processing unit 11, the recycling processing unit 13, or the waste processing unit 15. If the discrimination processing unit 45 determines that the object is not a textile product, the transport device 31 transports the object to either the recycling processing unit 13 or the waste processing unit 15. The discrimination processing unit 45 may be configured to determine whether the object is clothing using image data obtained by photographing the object. Alternatively, the discrimination processing unit 45 may be configured to determine whether the object is clothing using image data obtained by photographing the object.

[0033] A specific example will be described in which the discrimination processing unit 45 determines whether an object is clothing. Clothing can be classified into multiple categories based on its shape. The multiple categories include, for example, shirts, jackets, pants, and skirts. The discrimination processing unit 45 can also recognize the parts that make up clothing. For example, sleeves, collars, pockets, zippers, and hoods.

[0034] The sorting processing unit 37 is expected to be able to classify objects into categories and recognize the components that make up clothing by using image data obtained by photographing the objects with the camera 23. However, accurate classification is difficult based solely on visual information. For example, jeans are basically pants made of denim fabric, but similar colors, patterns, etc. can be reproduced by printing. Therefore, in this embodiment, the sensor 25 measures the weight of the object and the infrared sensor measures the material, etc. The sorting processing unit 37 judges the object based on the image data obtained by the camera 23 and the measurement data obtained by the sensor 25. The learning processing unit 43 performs a learning process to generate a model used by the sorting processing unit 37 for AI-based judgment.

[0035] The program for implementing the selection processing unit 37 and the like may be stored in advance in the information processing device and executed, or may be downloaded from another information processing device and executed. Also, it may be implemented using technology such as AI.

[0036] 2 is a flow chart showing an example of a learning process for generating a model used for determination by the discrimination processing unit 45. The learning process for generating a model will be described with reference to FIG.

[0037] The administrator of the sorting system 1 sends the teacher object to the unfolding device 21. The unfolding device 21 unfolds the teacher object. The camera 23 and the sensor 25 photograph and measure the teacher object unfolded by the unfolding device 21, respectively. The preprocessing analysis device 27 transmits the image data and measurement data obtained by the camera 23 and the sensor 25 to the sorting device 5.

[0038] The preprocessing result acquisition unit 39 receives the image data and measurement data obtained by the camera 23 and the sensor 25 from the preprocessing analysis device 27 .

[0039] The supervising processing unit 41 creates supervising data by adding annotation data corresponding to the image data and measurement data obtained by the camera 23 and the sensor 25. The annotation data is data indicating labels assigned to the data to be analyzed in supervised learning. The supervising processing unit 41 can create annotation data by referring to the data input into the sorting device 5 by the administrator of the sorting system 1 in correspondence with the image data and measurement data.

[0040] In this example, the annotation data includes whether or not the item is clothing.

[0041] In the case of clothing, the annotation data includes categories (shirt, jacket, pants, skirt, etc.) classified according to the overall shape of the clothing, parts (sleeves, collar, pockets, zippers, hood, etc.) classified according to the position and shape of the parts that make up the clothing, and accessories (buttons, etc.) present in each part. Furthermore, in the case of clothing, the annotation data includes data indicating the value of the clothing (for example, the value corresponding to the category, the position and shape of a mark indicating the brand and its corresponding value, etc.) and data indicating the value of the material (for example, the content of recyclable materials). The difficulty of reuse varies greatly depending on the category. For example, clothing that comes into direct contact with the skin is often difficult to reuse. On the other hand, clothing worn as a jacket is often easy to reuse. Therefore, the value of clothing varies greatly depending on the category.

[0042] The annotation data includes data indicating the value of the material if it is not clothing.

[0043] The administrator of the selection system 1 performs work to generate teacher data using multiple teacher objects. The teacher processing unit 41 generates multiple pieces of teacher data corresponding to the multiple teacher objects (step STA1 in FIG. 2). Note that for the same teacher object, different teacher data may be generated depending on whether the object is placed face up in the unfolding device 21 or face down.

[0044] The learning processing unit 43 performs a learning process (step STA2 in FIG. 2) using the plurality of teacher data generated by the teacher processing unit 41. The learning processing unit 43 acquires a model through the learning process (step STA3 in FIG. 2).

[0045] 3 and 4 are a block diagram and a flow diagram, respectively, for explaining the sorting process of collected items (clothes, etc.) in the sorting system 1. In Fig. 3, the same reference numerals are used to designate components corresponding to those in Fig. 1.

[0046] A user of the sorting system 1 carries collected items (collected clothing, etc.) into the unfolding device 21 (step STB1 in FIG. 4). The unfolding device 21 unfolds the collected items. The camera 23 and the sensor 25 photograph and measure the collected items unfolded by the unfolding device 21, respectively (step STB2 in FIG. 4). The unfolding device 21 sends the photographed and measured collected items to the sorting device 5. The pre-processing analysis device 27 transmits the image data and measurement data obtained by the camera 23 and the sensor 25 to the sorting device 5.

[0047] The preprocessing result acquisition unit 39 receives the image data and measurement data obtained by the camera 23 and the sensor 25 from the preprocessing analysis device 27 .

[0048] Using this model, the discrimination processing unit 45 uses the image data and measurement data obtained by the camera 23 and sensor 25 to determine whether the collected item sent by the unfolding device 21 to the sorting device 5 is clothing (step STB3). If the discrimination processing unit 45 determines that the collected item is clothing, the process proceeds to step STB4. If the discrimination processing unit 45 determines that the collected item is not clothing, the process proceeds to step STB9.

[0049] When the discrimination processing unit 45 determines that the collected item is clothing, it can use the model to obtain data indicating the category classified according to the overall shape of the clothing, the parts classified according to the positions and shapes of the parts that make up the clothing, the accessories (buttons, etc.) present on each part, etc. Furthermore, the discrimination processing unit 45 can use the model to obtain data indicating the value as clothing and the value as material.

[0050] Furthermore, when the discrimination processing unit 45 determines that the collected item is not clothing, it can use the model to obtain data indicating the value as a material.

[0051] In step STB4, the discrimination processing unit 45 determines whether the value of the clothing is equal to or greater than the first standard value. If the discrimination processing unit 45 determines that the value of the clothing is equal to or greater than the first standard value, the discrimination processing unit 45 sorts the collected items to the reuse processing unit. The transport device 31 transports the collected items to the reuse processing unit 11 (step STB5). If the discrimination processing unit 45 determines that the value of the clothing is less than the first standard value, the discrimination processing unit 45 determines whether the value of the material is equal to or greater than the second standard value (step STB6). If the discrimination processing unit 45 determines that the value of the material is equal to or greater than the second standard value, the discrimination processing unit 45 sorts the collected items to the recycling processing unit 13. The transport device 31 transports the collected items to the recycling processing unit 13 (step STB7). If the discrimination processing unit 45 determines that the value of the material is less than the second standard value, the discrimination processing unit 45 sorts the collected items to the waste processing unit 15. The transport device 31 transports the collected items to the waste processing unit 15 (step STB8).

[0052] In step STB9, the discrimination processing unit 45 determines whether the value as a material is equal to or greater than a third standard value. If the discrimination processing unit 45 determines that the value as a material is equal to or greater than the third standard value, the discrimination processing unit 45 sorts the collected item to the recycling processing device 13. The transport device 31 transports the collected item to the recycling processing device 13 (step STB10). If the discrimination processing unit 45 determines that the value as a material is less than the third standard value, the collected item is sorted to the waste processing device 15. The transport device 31 transports the collected item to the waste processing device 15 (step STB11).

[0053] FIG. 5 is (a) a block diagram showing an example of the configuration of the reuse processing device 11, and (b) a flow diagram showing an example of the operation thereof.

[0054] The reuse processing device 11 includes a proximity measurement device 51 and a value assessment device 53. The value assessment device 53 includes a proximity analysis processing unit 55 and a value assessment processing unit 57.

[0055] Color changes present in the collected clothing are observed in the image data captured by the camera 23. The proximity measurement device 51 photographs and / or measures, for example, the color changes present in the collected clothing from a closer distance than the photographing distance when the camera 23 photographs the collected clothing.

[0056] Color changes in the image data of recovered items can be due to prints or other factors that exist when the garment is made, or due to damage caused during use. Examples of the latter include stains (soiling caused by liquids or other factors) and scratches (damage caused by external forces).

[0057] The proximity analysis processor 55 uses the measurement data from the proximity measurement device 51 to analyze whether color changes in image data obtained by photographing the collected items were present when the clothing was made or whether they occurred after use. The proximity analysis processor 55 can be realized using technology such as AI. Using the measurement data from the proximity measurement device 51, the proximity analysis processor 55 identifies areas of the collected items that have been damaged during use (step STC1 in FIG. 5(b)).

[0058] The value assessment processing unit 57 assesses the resale value of the collected items by subtracting the value lost through use from the value as clothing.

[0059] The value of clothing varies depending on, for example, whether or not it has accessories. For example, if accessories are replaced, the value may increase or decrease depending on the material of the replaced accessories. Therefore, the value assessment processing unit 57 may assess the value of the clothing by analyzing the presence or absence of accessories and their material (gold, silver, etc.) using image data obtained by photographing the recovered item and measurement data from the proximity measurement device 51. Furthermore, for example, the value of clothing can be assessed by the discrimination processing unit 45 based on a mark indicating a brand, but the mark indicating the brand may be partially damaged due to use. In such cases, the value of the recovered item is reduced by the damaged portion. Therefore, the value assessment processing unit 57 assesses the value of the clothing by evaluating the change in value from the state when the clothing was made through analysis by the proximity analysis processing unit 55, and then modifies the value of the clothing assessed by the discrimination processing unit 45 based on the changed value to assess the value of the recovered item. The value assessment processing unit 57 can be implemented using technology such as AI, using image data obtained by photographing the recovered item, reference data on the damaged areas identified by the proximity analysis processing unit 55, and the like.

[0060] Furthermore, for example, the value deducted from the state in which the clothing was made will differ depending on the parts that the proximity analysis processing unit 55 assesses as damaged. The value assessment processing unit 57 assesses the damage to the clothing through analysis by the proximity analysis processing unit 55 and assesses the value of the recovered item.

[0061] If the reuse processing device 11 determines that the collected item cannot be reused, for example, because it is heavily damaged, it may send the collected item to the recycle processing device 13 or waste processing device 15.

[0062] FIG. 6 is a block diagram showing an example of the configuration of the recycle processing device 13 (a), and a flow chart showing an example of the operation (b).

[0063] The recycling processing device 13 includes a recycling measurement device 61, a collection preparation device 63, a fiber collection device 65, and a recycling analysis device 67. The recycling measurement device 61 includes a material measurement device 69 and an accessory measurement device 71. The recycling analysis device 67 includes a recycling analysis processing unit 73.

[0064] The material measuring device 69 performs a measurement process to identify the material of the fabric of each part of the collected item. The accessory measuring device 71 performs a measurement process for accessories (buttons, etc.) of the collected item.

[0065] The accessory measuring device 71 is capable of measuring the material of accessories, distinguishing them from fabric. Examples of measurable accessory materials include metal, pearl, silicon, and resin. Because accessories pose significant problems during processes such as cutting and recycling, they must be removed before recycling. However, accessories may be located in locations that are difficult to photograph with the camera 23, making it difficult to accurately identify their presence. On the other hand, measuring every part of a collected item is time-consuming. For example, there are parts of clothing that tend to be used with accessories, such as jeans, where both ends of the pockets are reinforced with metal rivets, and parts that tend not to be used with accessories. The discrimination processing unit 45 uses a model to obtain data indicating the accessories present for each part of the collected item. Therefore, the accessory measuring device 71 uses the data indicating the accessories present for each part of the collected item to perform measurement processing according to the accessory usage tendency for each part. For example, the accessory measuring device 71 uses the data indicating the accessories present for each part of the collected item to perform individual measurement processing for parts that tend to be used with accessories, while performing collective measurement processing for multiple parts that tend not to be used with accessories. The accessory measuring device 71 can perform highly accurate measurement processing on the accessory.

[0066] An example of the operation of the recycling processing device 13 will be described with reference to FIG. 6(b). The recycling measurement device 61 performs measurement processing on the collected items (step STD1). The recycling analysis processing unit 73 analyzes the fabric material and accessories of the collected items using image data obtained by photographing the collected items and measurement data from the recycling measurement device 61 (step STD2). The recycling analysis processing unit 73 can be implemented using technology such as AI. The collection preparation device 63 uses the analysis results from the recycling analysis processing unit 73 to remove accessories from the items and cut the fabric (step STD3). The fiber recovery device 65 processes the cut fabric according to the material and recovers fibers (step STD4). The process for determining the material in the fiber recovery device 65 can be implemented using technology such as AI.

[0067] If the recycling processing device 13 determines that the collected item cannot be recycled, it may send the collected item to the waste processing device 15.

[0068] FIG. 7 is a block diagram showing an example of the configuration of the waste processing device 15 (a) and a flow chart showing an example of the operation (b).

[0069] The waste processing device 15 includes a waste measurement device 81, a waste device 83, and a waste evaluation device 85. The waste evaluation device 85 includes a waste analysis processing unit 87.

[0070] The disposal measuring device 81 performs measurement processing for disposing of the collected items. The disposal analysis processing unit 87 uses the measurement data from the disposal measuring device 81 to perform analysis for disposing of the collected items (step STE1 in FIG. 7(b)). The disposal analysis processing unit 87 analyzes, for example, whether the collected items are combustible or non-combustible. The disposal analysis processing unit 87 can be realized by technology such as AI. The disposal device 83 performs disposal processing using the analysis results of the disposal analysis processing unit 87 (step STE2 in FIG. 7(b)).

[0071] An example of a specific configuration of the unfolding device 21, camera 23, and sensor 25 in the pre-processing device 3 of FIG. 1 will be described with reference to FIG.

[0072] In the collected state, multiple pieces of clothing are bunched together and present in a clump. FIG. 8 shows an example of an apparatus for separating multiple pieces of clothing present in a clump and removing them as individual collected items. In this example, the spreading device (an example of the spreading device 21 in FIG. 1) includes a lower conveying device and a third conveying device 105. The lower conveying device includes a first conveying device 101 and a second conveying device 103. The first conveying device 101, the second conveying device 103, and the third conveying device 105 can be realized, for example, by belt conveyors. In the lower conveying device, the first conveying device 101 and the second conveying device 103 rotate the belts clockwise to convey clothing and other items present on the upper surface from left to right in the figure. The third conveying device 105 is an upper conveying device that exists above at least a portion of the lower conveying device, and is higher on the left (the starting point side of the conveying direction by the lower conveying device) and lower on the right (the end point side of the conveying direction by the lower conveying device). The third conveying device 105 rotates the belt clockwise, pulling the uppermost clothes in a pile of clothes being conveyed by the lower conveying device in the opposite direction to the conveying direction of the lower conveying device, allowing the lower conveying device to separate and convey the lowermost clothes. The third conveying device 105 can also be realized by a robot arm or the like.

[0073] FIG. 8(b) shows a state in which clothes 107 and clothes 109 are present on the second conveyance device 103. The first camera 111 (an example of the camera 23 in FIG. 1) captures images of the clothes 107 and clothes 109 present on the second conveyance device 103. The inventors have experimentally confirmed that good image capture is possible when the distance between the second conveyance device 103 and the first camera 111 is approximately 1 to 2 m. The first camera 111 may capture still images or videos. The weight sensor 113 (an example of the sensor 25 in FIG. 1) is a sensor that measures the weight of the clothes 107 and clothes 109 present on the second conveyance device 103. For example, even if multiple clothes 107 and clothes 109 are present, the weight of each of the clothes 107 and clothes 109 can be measured from the image data obtained by the first camera 111 by analyzing the increase and decrease over time.

[0074] Another example of the specific configuration of the unfolding device 21, camera 23, and sensor 25 in the pre-processing device 3 of FIG. 1 will be described with reference to FIGS.

[0075] In this example, the spreading device (an example of the spreading device 21 in FIG. 1) includes a fourth conveying device 121, a fifth conveying device 123, a sixth conveying device 125, and a seventh conveying device 127. The fourth conveying device 121, the fifth conveying device 123, the sixth conveying device 125, and the seventh conveying device 127 can be realized by, for example, belt conveyors. This example can be positioned as, for example, the second conveying device 103 in the example of FIG. 8.

[0076] When garments 129 are individually carried in from outside, the fourth conveying device 121 rotates the belt clockwise to convey one side (for example, the front) of the garment 129 facing up. The second camera 131 (an example of the camera 23 in FIG. 1) photographs the top side of the garment 129 being conveyed by the fourth conveying device 121.

[0077] The fifth conveying device 123 conveys the garments 129 conveyed by the fourth conveying device 121 with the bottom side (e.g., the back body) facing up. The sixth conveying device 125 also conveys the garments 129 conveyed by the fifth conveying device 123. As a result, the fifth conveying device 123 and the sixth conveying device 125 turn the garments 129 upside down, allowing the third camera 133 (an example of camera 23 in FIG. 1) to capture the bottom side that could not be captured by the second camera 131, and thereby capturing images of both sides of the garments 129. The shear sensors 1351 and 1352 (an example of sensor 25 in FIG. 1) grasp and fix the garments 129 in the fifth conveying device 123 and the sixth conveying device 125, respectively, and measure the shear force of the garments 129 by moving at least one of the fifth conveying device 123 and the sixth conveying device 125 away from each other.

[0078] The seventh conveying device 127, following the sixth conveying device 125, conveys the garment 129 conveyed by the sixth conveying device 125. FIG. 10 shows an example of the configuration of the seventh conveying device 127 and near-infrared sensors 1371 and 1372. The seventh conveying device 127 includes two belts 1391 and 1392. The seventh conveying device 127 conveys the garment 129 by the two belts 1391 and 1392. The near-infrared sensors 1371 and 1372 (an example of sensor 25 in FIG. 1) are sensors that perform measurements using near-infrared rays, and are positioned and move between the two belts 1391 and 1392 to measure multiple locations on the garment 129. Note that the present invention may also be understood as a device in which an unfolding device includes a conveying device that conveys an object, the conveying device uses multiple conveying components for conveyance, and the conveying device includes sensors that move between the multiple conveying components to measure multiple locations on the object.

[0079] FIG. 11 shows a camera (an example of a combination of the camera 23 and discrimination processing unit 45 in FIG. 1) incorporating image recognition AI (artificial intelligence) developed by the inventor. At present, processing can be completed in less than one second. By incorporating AI into the camera, it is possible to specialize it for specific processing and perform image recognition in parallel. For example, it is possible to improve the accuracy of judgment by taking pictures from multiple directions, and to improve the accuracy of image recognition by taking pictures at different heights.

[0080] 12 shows an example of the discrimination result by the discrimination processing unit 45. The category has been determined to be a jacket 141. In addition, the parts have been determined to be a lapel (the lower collar of the jacket) 143, sleeves 145 and 147, and pockets 149 and 151.

[0081] 13 shows an example of the result of accessory discrimination by the discrimination processing unit 45. In FIGS. 13(a) and 13(b), metal buttons 153 and 155 can be discriminated.

[0082] The inventors have confirmed through experiments that the discrimination processing unit 45 can recognize brand marks.

[0083] Fig. 14 is a diagram showing an example of the proximity measurement device 51 in Fig. 5. In this example, a proximity camera 157 was brought into close proximity and moved over the clothing, capturing multiple images. It was confirmed that the proximity analysis processing unit 55 could identify scratches and stains on the clothing from the multiple images.

[0084] Figures 15 and 16 show examples of material analysis by the recycling analysis device 67 of Figure 6. The recycling analysis processing unit 73 can be realized by, for example, a processor that operates under the control of a program.

[0085] The recycle analysis processing unit 73 first performs a primary determination by using multiple analyses to determine whether the material is similar to multiple target materials (e.g., cotton, polyester) or not. Figure 15 shows the analysis results for cotton, linen, wool, silk, acrylic, polyester, polyurethane, nylon, rayon, cupra, and acetate obtained through two analyses. In this way, by using multiple analyses, the target materials can be divided into groups. If the distance between the material analysis results and each group is less than a reference value, the recycle analysis processing unit 73 can determine that the material is similar to at least one group. If the distance between the material analysis results and each group is equal to or greater than a reference value, the recycle analysis processing unit 73 can determine that the material is not similar.

[0086] If the analysis results of the material are similar to at least one of the target materials, the recycle analysis processor 73 performs a secondary judgment by estimating the ratio of the target material using the results of principal component analysis (PCA) and estimating which of the target materials the material is similar to. Figure 16 shows the results of a detailed analysis of cotton and polyester. The results show 11 principal component analyses performed with different ratios of cotton and polyester. The materials are grouped based on their ratios. Therefore, if the distance between the material analysis results and each group is less than the reference value, the recycle analysis processor 73 can estimate the ratio of cotton to polyester, as it is close to at least one group. Similar estimations are possible for other ratios of target materials (cotton, linen, wool, silk, acrylic, polyester, polyurethane, nylon, rayon, cupra, acetate, etc.). It is preferable to use AI for the secondary judgment by the recycle analysis processor 73.

[0087] The fiber recovery device 65 of FIG. 6 can recover fibers by performing recovery processing according to the detected material to be determined using the analysis results. [Explanation of symbols]

[0088] 1. Sorting system 3 Pretreatment equipment 5. Sorting equipment 11 Reuse processing equipment 13 Recycling processing equipment 15 Waste treatment equipment 21 Deployment device 23 Camera 25 sensors 27 Pretreatment analysis equipment 31 Transport device 33 Sorting control device 35 Selection memory unit 37 Sorting processing section 39 Preprocessing result acquisition unit 41 Teacher Processing Section 43 Learning processing unit 45 Discrimination processing unit 51 Proximity Measuring Device 53 Value Assessment Device 55 Proximity Analysis Processing Unit 57 Value Assessment Processing Unit 61 Recycling Measuring Device 63 Recovery preparation device 65 Fiber recovery equipment 67 Recycling analysis equipment 69 Material Measuring Device 71 Accessory measuring device 73 Recycling analysis processing unit 81 Disposal Measuring Device 83 Disposal Device 85 Disposal evaluation device 87 Waste Analysis Processing Unit 101 First conveying device 103 Second conveying device 105 Third conveying device 107 Clothes 109 Clothes 111 Camera 1 113 Weight Sensor 121 Fourth transport device 123 5th Transport Device 125 6th Transport Device 127 7th Transport Device 129 Clothes 131 Second Camera 133 Third Camera 135 Shear force sensor 137 Near-infrared sensor 139 Belt 141 Jacket 143 Lapel (Jacket collar) 145 sleeve 147 Sleeve 149 pockets 151 pockets 153 Metal Buttons 155 Metal Buttons 157 Close-up Camera

Claims

1. A sorting system for sorting collected items, The system comprises a sorting device, a reuse processing device that evaluates the resale value of the collected items, a recycling processing device that recovers fibers from the collected items, and a waste processing device, The sorting device includes a conveying device and a discrimination processing unit, the discrimination processing unit determines whether the collected item is a textile product using image data obtained by photographing the collected item, The conveying device is If the discrimination processing unit determines that the collected product is a textile product, the collected product is transported to one of the reuse processing unit, the recycle processing unit, and the waste processing unit; If the discrimination processing unit determines that the collected item is not a textile product, the sorting system transports the collected item to either the recycling processing unit or the waste processing unit.

2. the reuse processing device includes a proximity measurement device and a value assessment device, The value assessment device includes a proximity analysis processing unit and a value assessment processing unit, the proximity measurement device measures the collected items at a distance closer than the distance at which the collected items were photographed to obtain the image data; the proximity analysis processing unit uses the measurement data of the proximity measurement device to identify a portion of the recovered item that has been damaged during use; The sorting system according to claim 1 , wherein the value assessment processing unit assesses the value of the recovered items by referring to the damaged portion identified by the proximity analysis processing unit.

3. the discrimination processing unit determines whether the collected item is clothing using the image data; The recycling processing device includes an accessory measuring device and a collection preparation device, When the discrimination processing unit determines that the collected item is clothing, it obtains data indicating accessories present for each part of the collected item, The accessory measuring device is capable of measuring the material of the accessory separately from the fabric, the accessory measuring device performs a measurement process according to the tendency of accessories to be used for each part using data indicating accessories present for each part of the recovered goods; The sorting system according to claim 1 , wherein the collection preparation device removes accessories detected by the measurement of the accessory measuring device from the collected items.

4. The recycling processing device includes a material measuring device, a recycling analysis device, a recovery preparation device, and a fiber recovery device, The material measuring device measures the material of the recovered items, The recycling analysis device is As a first step, we use multiple analyses to determine whether the material is similar to multiple target materials. If the analysis result of the material is close to at least one of the materials to be judged, the ratio of the material to be judged is estimated as a secondary judgment, The collection preparation device cuts the collected items, The sorting system according to claim 1 , wherein the fiber recovery device performs a fiber recovery process on the cut collected items in accordance with the detected target material.

5. A pretreatment device is provided, The pre-treatment device includes a deployment device, a camera, and a sensor; The deployment device deploys the retrieved items, the camera photographs the recovered items deployed by the deployment device to obtain the image data; The sensor measures the retrieved item deployed by the deployment device; 2. The sorting system of claim 1, wherein the discrimination processing unit uses the image data and measurement data obtained by the camera and the sensor to determine whether the collected item is clothing, and if it determines that the collected item is clothing, obtains data indicating the category into which the clothing is classified, the parts that make up the clothing, and the accessories present for each of the parts.

6. The unfolding device includes a lower conveying device and an upper conveying device, the upper conveying device is located above the lower conveying device, The sorting system according to claim 5 , wherein the upper conveying device conveys the uppermost of the piled collected products in a direction opposite to the direction in which the lower conveying device conveys the collected products.

7. A reuse processing device that evaluates the resale value of collected items, A proximity measuring device and a value evaluating device are provided, The value assessment device includes a proximity analysis processing unit and a value assessment processing unit, the proximity measurement device measures the collected item at a distance closer than the distance at which the collected item was photographed to obtain image data; the proximity analysis processing unit uses the measurement data of the proximity measurement device to identify a portion of the recovered item that has been damaged during use; The value assessment processing unit assesses the value of the recovered item by referring to the image data and the damaged portion identified by the proximity analysis processing unit.

8. A recycling processing device that recovers fibers from recovered items, The apparatus includes a material measuring device, an accessory measuring device, a recovery preparation device, and a fiber recovery device, If the collected item is determined to be clothing, data indicating accessories present for each part of the collected item is generated; The material measuring device measures the material of the recovered items, The accessory measuring device is capable of measuring the material of the accessory separately from the fabric, the accessory measuring device performs a measurement process according to the tendency of accessories to be used for each part using data indicating accessories present for each part of the recovered goods; the collection preparation device removes the accessories detected by the measurement of the accessory measuring device from the collected items; The collection preparation device cuts the collected items, The fiber recovery device is a recycling processing device that performs fiber recovery processing on the cut collected items in accordance with the detected material to be determined.

9. 1. A waste treatment device, comprising: The sorting device is equipped with a disposal device that disposes of collected items that the sorting device determines cannot be reused or recycled. The sorting device is a waste processing device that determines whether the collected items can be reused or recycled, or both, based on image data and measurement data obtained by using a camera and sensor to photograph and measure the items.

10. A sorting method in a sorting system for sorting recovered items, comprising: the sorting system includes a sorting device, a reuse processing device that evaluates the resale value of the collected items, a recycling processing device that recovers fibers from the collected items, and a waste processing device; The sorting device includes a conveying device and a discrimination processing unit, a step in which the discrimination processing unit determines whether the collected item is a textile product using image data obtained by photographing the collected item; The conveying device If the discrimination processing unit determines that the collected product is a textile product, the collected product is transported to one of the reuse processing unit, the recycle processing unit, and the waste processing unit; The sorting method includes a step of transporting the collected items to either the recycling processing device or the waste processing device if the discrimination processing device determines that the collected items are not textile products.

11. A program for causing a computer to function as the discrimination processing unit according to claim 1.

Images