Identification device, classification device, identification method, classification method, and item manufacturing method

Abstract

This identification device identifies a specific type of object from among a plurality of objects being conveyed on a conveyance path and comprises: a detection unit that detects the size of each of a plurality of objects including a first object and a second object in a detection region in the conveyance path; a measurement unit that irradiates an object selected from among the plurality of objects with light in a measurement region on a downstream side of the detection region in the conveyance path, and measures reflected light from the object; and a control unit that identifies the specific type of object on the basis of a measurement result of the measurement unit. When it is determined that both the first object and the second object cannot be measured, the control unit determines an object to be measured by the measurement unit on the basis of the size of the first object and the second object detected by the detection unit.

Description

Identification device, classification device, identification method, classification method, and article manufacturing method

[0001] The present disclosure relates to an identification device, a classification device, an identification method, a classification method, and an article manufacturing method.

[0002] Conventionally, industrial waste such as automobiles and home appliances has been separated into metals such as iron and aluminum and various types of resins (plastics) in various recycling processes. In recent years, attempts have been made to reuse these resins as materials for resin molded parts through horizontal recycling, and systems have been developed that identify and classify specific types (materials) of resin from various types of resin. Patent Document 1 discloses an apparatus that sequentially performs an identification process on multiple objects transported on a belt conveyor, measuring the reflected light from the objects (resins) upon irradiating them with light and identifying the type of the objects.

[0003] JP 2023-167533 A

[0004] In a configuration in which a recognition process is sequentially performed on a plurality of objects being transported on a belt conveyor (transport path) like the device described in Patent Document 1, if the plurality of objects are arranged densely, the recognition process may not be performed on some of the plurality of objects. In this case, if the recognition process is not performed on relatively large objects, the recovery rate of a particular type of object may decrease.

[0005] The present disclosure provides a technique that is advantageous for recovering a specific type of object from a plurality of objects being transported on a transport path.

[0006] An identification device according to one aspect of the present disclosure is an identification device that identifies a specific type of object from among a plurality of objects being transported on a transport path, and includes: a detection unit that detects the size of each of a plurality of objects, including a first object and a second object, in a detection area of ​​the transport path; a measurement unit that irradiates light onto an object selected from the plurality of objects and measures the light reflected from the object in a measurement area of ​​the transport path downstream of the detection area; and a control unit that identifies the specific type of object based on the measurement results of the measurement unit, wherein if the control unit determines that it is unable to measure both the first object and the second object, it determines the object to be measured by the measurement unit based on the sizes of the first object and the second object detected by the detection unit.

[0007] According to the present disclosure, for example, it is possible to provide a technique that is advantageous in recovering a specific type of object from a plurality of objects being transported on a transport path.

[0008] Other features and advantages of the present disclosure will become apparent from the following description taken in conjunction with the accompanying drawings, in which the same or similar components are designated by the same reference numerals.

[0009] The accompanying drawings are included in the specification, constitute a part thereof, illustrate embodiments of the present disclosure, and are used together with the description to explain the principles of the present disclosure. A schematic diagram showing an example of the configuration of a classification device according to a first embodiment. A diagram schematically showing an example of measurement processing by a measurement unit for a plurality of resin pieces being transported on a belt conveyor. A diagram schematically showing an example of measurement processing by a measurement unit for a plurality of resin pieces being transported on a belt conveyor. A flowchart showing the operation flow of classification processing according to the first embodiment. A flowchart showing the operation flow of selection processing and measurement processing according to the first embodiment. A diagram showing an example of selecting target resin pieces using a beam search method. A flowchart showing the operation flow of identification processing according to a second embodiment. A diagram showing an example of a function expressing a change in value threshold value relative to an intrusion interval. A flowchart showing the operation flow of selection processing and measurement processing according to the second embodiment. A diagram showing another example of a function expressing a change in value threshold value relative to an intrusion interval. A diagram showing another example of a function expressing a change in value threshold value relative to an intrusion interval. A diagram showing an example of changing a function expressing a change in value threshold value relative to an intrusion interval according to measurement time.

[0010] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the claims. Although multiple features are described in the embodiments, not all of these multiple features are required, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same reference numerals are used to designate the same or similar components, and redundant explanations will be omitted.

[0011] First Embodiment A classification device 100 (classification system) according to a first embodiment of the present disclosure will be described. The classification device 100 is a device that classifies a plurality of objects being transported on a transport path. Specifically, the classification device 100 identifies a specific type of object from among a plurality of objects being transported on the transport path, and selects the specific type of object based on the identification result.

[0012] In this embodiment, the objects transported along the transport path will be described using crushed pieces of plastic resin material as an example. The plastic resin material is, for example, a thermoplastic plastic or a thermosetting plastic, and is crushed into pieces of approximately 10 to 100 mm in size before being transported along the transport path. Hereinafter, crushed pieces of plastic resin material may be referred to as "resin pieces."

[0013] This embodiment also describes an example of sorting resin pieces as objects using a known Raman spectroscopy technique. When excitation light is irradiated onto a resin piece, the wavelength of the reflected light shifts due to Raman spectroscopy. The amount of wavelength shift varies depending on the material of the resin piece. Therefore, the wavelength distribution of the reflected light to be obtained from the material to be identified is acquired in advance as reference data (teaching data). By comparing the reference data with the measured data of the reflected light from the resin piece, the material (component) of the resin piece can be identified. Examples of materials that can be identified include ABS resin, polypropylene (PP), and polyethylene (PE). Using this technique, a specific type of resin piece that a user wants to sort can be identified from multiple resin pieces transported on a transport path, and the specific type can be sorted using an air jet or the like.

[0014] 1 is a schematic diagram showing an example configuration of a sorting device 100 (sorting system) according to this embodiment. The sorting device 100 may include a belt conveyor 10, a detection unit 20, a measurement unit 30, a control unit 40, and a sorting unit 50. In the following description, directions are indicated by an XYZ coordinate system in which the surface (top surface 11) of the belt conveyor 10 on which the multiple resin pieces SP are placed is defined as the XY plane, and the direction in which the multiple resin pieces SP are transported by the belt conveyor 10 is defined as the Y direction.

[0015] The belt conveyor 10 is a transport mechanism that transports multiple resin pieces SP on a transport path, and the transport path is defined by an upper surface 11 of the belt conveyor 10. Multiple resin pieces SP are randomly fed onto the upper surface 11 of the belt conveyor 10 via a fixed-amount cutting device (crusher) or a vibrating feeder (not shown). Here, the belt conveyor 10 moves a belt made of rubber, resin, or metal at a predetermined speed, and the size and moving speed of the belt can be set according to the processing capacity of the sorting device 100.

[0016] The detection unit 20 detects the size (dimension) of each of the multiple resin pieces SP in a detection area 20' of the conveying path. For example, the detection unit 20 includes an imaging unit 21 having a detection area 20' (imaging field of view) that is a portion of the upper surface 11 of the belt conveyor 10, and a processing unit 22 that determines the size of each resin piece SP by performing predetermined image processing on images obtained by the imaging unit 21. The detection area 20' is set to a shape extending in the width direction (X direction) of the belt conveyor 10, and the imaging unit 21 may be configured to continuously image each resin piece SP being conveyed by the belt conveyor 10. The imaging unit 21 may be configured as an area camera capable of acquiring three-dimensional images or as a line camera capable of acquiring two-dimensional images. This allows the processing unit 22 to determine the size (XY direction) of each resin piece SP based on the multiple images continuously obtained by the imaging unit 21. The detection unit 20 may have a sensor, instead of the imaging unit 21, for detecting the presence or absence of the resin piece SP within the detection area 20' (that is, the passage of the resin piece SP through the detection area 20').

[0017] The detection unit 20 can also detect the position of each resin piece SP on the upper surface 11 (on the conveying path) of the belt conveyor 10. The detection area 20' is set to a shape extending in the width direction (X direction) of the belt conveyor 10. Therefore, the processing unit 22 can determine the position of each resin piece SP in both the conveying direction (Y direction) and the width direction (X direction) of the belt conveyor 10 based on the images obtained by the imaging unit 21. In this embodiment, the processing unit 22 can determine the leading edge position of each resin piece SP in the conveying direction as the position of each resin piece SP in the conveying direction. Furthermore, the detection unit 20 can detect the brightness of each resin piece SP on the upper surface 11 (on the conveying path) of the belt conveyor 10. For example, the processing unit 22 can determine the brightness of each resin piece SP based on the contrast of the image obtained by the imaging unit 21.

[0018] The measurement unit 30 performs a measurement process in a measurement region 30′ downstream of the detection region 20′ of the conveyance path, irradiating the resin piece SP conveyed on the belt conveyor 10 with irradiation light 31 (excitation laser light) and measuring the light reflected from the resin piece SP. The measurement unit 30 may include a scanning mechanism 32 that scans the irradiation light 31 in the measurement region 30′ and a spectroscope 33 that disperses (measures) the Raman scattered light from the resin piece SP as reflected light. The scanning mechanism 32 may be configured, for example, by a galvanometer scanner and may be controlled by the control unit 40. Based on the detection results of the detection unit 20, the control unit 40 controls the scanning of the irradiation light 31 by the scanning mechanism 32 in accordance with the movement of the resin piece SP so that the irradiation light 31 is irradiated onto the resin piece SP moving in the conveyance direction within the measurement region 30′ for a predetermined time. The spectroscope 33 also disperses the reflected light into wavelengths, generates measurement data indicating the wavelength distribution of the reflected light, and supplies the measurement data to the control unit 40. Here, the measurement process by the measurement unit 30 can be performed in order for each resin piece SP being transported on the belt conveyor 10 in the measurement area 30'.

[0019] The control unit 40 is configured by a computer (information processing device) having a processor such as a CPU (Central Processing Unit) and a storage unit such as a memory, and controls the classification process in the classification device 100. In this embodiment, the control unit 40 selects a resin piece SP from among the multiple resin pieces SP being transported on the belt conveyor 10 based on the detection results of the detection unit 20, for which the measurement unit 30 will measure reflected light. The control unit 40 then causes the measurement unit 30 to measure the reflected light from the selected resin piece SP, and performs an identification process to identify a specific type of resin piece SP based on the measurement results. The control unit 40 may also have a user interface for receiving instruction input from a user.

[0020] Here, the detection unit 20, measurement unit 30, and control unit 40 in the sorting device 100 can constitute an identification device that identifies a specific type of resin piece SP from among multiple resin pieces SP being transported on the transport path by the belt conveyor 10. Hereinafter, the device including the detection unit 20, measurement unit 30, and control unit 40 may be referred to as the "identification device," and the processing performed by the identification device may be referred to as the "identification processing."

[0021] The sorting unit 50 (sorting mechanism) is disposed near the end of the belt conveyor 10 and sorts out specific types of resin pieces SP from among the multiple resin pieces SP transported by the belt conveyor 10. For example, the sorting unit 50 has multiple openings 51 (air nozzles) arranged along the width direction of the belt conveyor 10 and is configured as an air jet that releases compressed air from each opening 51 at an independent timing. This makes it possible to adjust the drop position for each resin piece SP transported by the belt conveyor 10.

[0022] As an example, under the control of the control unit 40, the sorting unit 50 releases compressed air from at least one opening 51 toward a resin piece SP identified as a specific type at the position and timing at which the resin piece SP is transported by the belt conveyor 10. As a result, the resin piece SP identified as a specific type is collected (stored) in the collection box 62, and the other resin pieces SP are collected (stored) in the collection box 61 by free fall. In other words, the specific type of resin piece SP is selected from the multiple resin pieces SP transported by the belt conveyor 10. The collection box 62 can be located farther from the belt conveyor 10 than the collection box 61.

[0023] In the above example, the sorting unit 50 discharges compressed air toward the specific type of resin pieces SP. However, the sorting unit 50 may discharge compressed air toward resin pieces SP other than the specific type. In this case, the specific type of resin pieces SP are collected in the collection box 61 by free fall, and the other resin pieces SP are collected in the collection box 62. The material, size, shape, and number of collection boxes can be set as desired depending on the configuration of the sorting device 100. For example, if there are multiple specific types of resin pieces SP to be sorted, the number of collection boxes is not limited to two, and may be three or more. In this case, the air volume (air speed) of the compressed air discharged from the sorting unit 50 can be controlled.

[0024] In the sorting device 100 (identification device) configured as described above, the measurement process by the measurement unit 30 is performed in order on each resin piece SP being transported on the belt conveyor 10. Since the measurement process by the measurement unit 30 takes a certain (reasonable) time, if multiple resin pieces SP are placed randomly on the belt conveyor 10 and arranged densely, the measurement process may not be performed on some of the multiple resin pieces SP. In this case, if the measurement process is not performed on relatively large resin pieces SP, the recovery rate of specific types of resin pieces SP may decrease.

[0025] Therefore, the control unit 40 of this embodiment selects resin pieces SP to be measured from the multiple resin pieces SP being transported on the belt conveyor 10 so that the larger the size detected by the detection unit 20, the higher the measurement process priority. The control unit 40 then performs measurement processing on the selected resin pieces SP using the measurement unit 30, and identifies specific types of resin pieces SP based on the measurement results. This makes it possible to avoid not performing measurement processing on relatively large resin pieces SP of a specific type, and improve the recovery rate of specific types of resin pieces SP.

[0026] 2A and 2B are views of the belt conveyor 10 viewed from above, and schematically show an example of measurement processing by the measurement unit 30 on multiple resin pieces SP being transported on the belt conveyor 10. FIG. 2A shows a conventional example, and FIG. 2B shows this embodiment. In addition, in FIGS. 2A and 2B, resin pieces SP that were targeted for measurement by the measurement unit 30 within the measurement area 30' are shown hatched, and resin pieces SP that were not targeted for measurement are shown in white. The dashed arrows in FIGS. 2A and 2B indicate the order (tracking) in which measurement processing is performed.

[0027] In the conventional example, as shown in FIG. 2A , the measurement process for the resin pieces SP was performed by the measurement unit 30 in the order in which they entered the measurement area 30′, without taking into account the size of the resin pieces SP. Therefore, the measurement process may not be performed on relatively large resin pieces SPa. In this case, even if the resin pieces SPa are of a specific type, they may not be identified as such, which may reduce the recovery rate of the specific type of resin pieces SP. In contrast, in the present embodiment, as shown in FIG. 2B , the order in which the measurement process is performed is determined taking into account the size of the resin pieces SP, and the measurement process is performed preferentially on relatively large resin pieces SPa. Therefore, it is possible to efficiently identify specific types of resin pieces SP from among multiple resin pieces SP, thereby improving the recovery rate of the specific type of resin pieces SP.

[0028] [Operation Flow of Classification Processing] The operation flow of the classification processing in the classification device 100 of this embodiment will be described below. Fig. 3 is a flowchart showing the operation flow of the classification processing of this embodiment. The flowchart of Fig. 3 is started when the main power of the classification device 100 is turned on, and can be executed by the control unit 40. Note that the classification device 100 includes an identification device, and the classification processing includes an identification processing.

[0029] In step S101, the control unit 40 performs a preparation process in response to receiving the preparation signal. The preparation process may include, for example, supplying power to the various units (belt conveyor 10, detection unit 20, measurement unit 30, and sorting unit 50) to activate the various units and make them ready to perform the classification process. The preparation signal is an instruction signal input by the user via the user interface to perform the preparation process (i.e., to activate the various units).

[0030] In step S102, the control unit 40 determines whether a stop signal has been received. A stop signal is an instruction signal input by a user via a user interface to stop the sorting device 100 when a problem occurs in the device (system) or when the user wants to end the sorting process. If a stop signal has been received, the process proceeds to step S111, where the control unit 40 stops the various units (belt conveyor 10, detection unit 20, measurement unit 30, and sorting unit 50) by cutting off power to those units. On the other hand, if a stop signal has not been received, the process proceeds to step S103.

[0031] In step S103, the control unit 40 determines whether or not a detection start signal has been received. The detection start signal is an instruction signal input by the user via the user interface to cause the detection unit 20 to start detecting each resin piece SP within the detection area 20'. If the detection start signal has not been received, the process returns to step S102; if the detection start signal has been received, the process proceeds to step S104.

[0032] In step S104, the control unit 40 determines whether the detection process by the detection unit 20 has already been performed. If the detection process has not yet been performed, the process proceeds to step S105, where the control unit 40 prepares the detection unit 20 for detection. For example, as the detection preparation for the detection unit 20, the control unit 40 causes the imaging unit 21 of the detection unit 20 to start capturing images of each resin piece SP in the detection area 20'. Then, in step S106, the control unit 40 causes the processing unit 22 of the detection unit 20 to start a detection process that detects (calculates) the size, position, and brightness of each resin piece SP based on the image obtained by the imaging unit 21. On the other hand, if the detection process has already been performed, the process skips steps S105 and S106 and proceeds to step S107.

[0033] In step S107, the control unit 40 determines whether or not a measurement start signal has been received. The measurement start signal is an instruction signal input by the user via the user interface to cause the measurement unit 30 to start measuring each resin piece SP within the measurement area 30'. If the measurement start signal has not been received, the process returns to step S102; if the measurement start signal has been received, the process proceeds to step S108.

[0034] In step S108, the control unit 40 determines whether the measurement process by the measurement unit 30 has already been performed. If the measurement process has not yet been performed, the process proceeds to step S109, where the control unit 40 prepares the measurement unit 30 for measurement. For example, as preparation for measurement, the control unit 40 activates the scanning mechanism 32 (galvanometer scanner) and performs preliminary oscillation of the irradiation light 31 (excitation light laser) so that the irradiation light 31 can be immediately turned on. Then, in step S110, the control unit 40 starts an identification process for identifying a specific type of resin piece SP based on the measurement results of the measurement unit 30. On the other hand, if the measurement process has already been performed, the process skips steps S109 and S110 and proceeds to step S102.

[0035] [Operation Flow of Selection Processing and Measurement Processing] The operation flow of the selection processing and measurement processing will be described below. Fig. 4 is a flowchart showing the operation flow of the selection processing and measurement processing of this embodiment. The flowchart of Fig. 4 starts in a state where the detection unit 20 and the measurement unit 30 are operating, and ends when step S111 of the flowchart of Fig. 3 is executed. Furthermore, during execution of the flowchart of Fig. 4, images of the detection area 20' can be continuously taken by the imaging unit 21 of the detection unit 20.

[0036] Here, the selection process refers to a process of selecting a target resin piece SP (hereinafter sometimes referred to as a target resin piece SPt) to be subjected to measurement processing from among the multiple resin pieces SP being transported on the belt conveyor 10 based on the detection results of the detection unit 20. The selection process may include steps S201 to S208 in the flowchart of FIG. 4. The measurement process refers to a process of measuring reflected light from the target resin piece SPt selected by the selection process using the measurement unit 30. The measurement process may include step S209 in the flowchart of FIG. 4.

[0037] Steps S201 to S202 are steps performed by the processing unit 22 of the detection unit 20. Note that if the processing unit 22 of the detection unit 20 is configured as part of the control unit 40, steps S201 to S202 may be performed by the control unit 40. In this case, the detection unit 20 can be configured by the imaging unit 21 and a part of the control unit 40 that has the function of the processing unit 22.

[0038] In step S201, the processing unit 22 determines whether or not a resin piece SP is present in the image obtained by the imaging unit 21. For example, the processing unit 22 can identify the resin piece SP in the image by performing known image processing on the image obtained by the imaging unit 21. Multiple resin pieces SP may be present in the image, and in this case, the processing unit 22 can identify each of the multiple resin pieces SP in the image. If no resin piece SP is present in the image, step S201 is repeated, and if it is determined that a resin piece SP is present in the image, the processing unit 22 proceeds to step S202.

[0039] In step S202, the processing unit 22 calculates (determines) the size, position, and brightness of each resin piece SP identified in step S201. For example, the processing unit 22 can calculate the size, position, and brightness of each resin piece SP in the image obtained by the imaging unit 21 by performing known image processing on the image. The size of the resin piece SP may be understood, for example, as the area in the XY plane. The position of the resin piece SP may include, for example, the position of the tip of the resin piece SP in the conveying direction (Y direction) and may also include the position (e.g., the center of gravity position) in the width direction (X direction) of the belt conveyor 10. The brightness of the resin piece SP may be understood, for example, as corresponding to the reflectance (color) of the resin piece SP. If it is difficult to calculate the size, etc. of the resin piece SP from a single image obtained by the imaging unit 21, the processing unit 22 may calculate the size, etc. of the resin piece SP from multiple images obtained consecutively by the imaging unit 21. Information indicating the size, position, and brightness of the resin piece SP calculated in step S202 is sent to the control unit 40.

[0040] In this embodiment, an example will be described in which the size, position, and brightness of the resin piece SP are all detected by the detection unit 20, but this is not limiting, and for example, only the size of the resin piece SP may be detected by the processing unit 22. Alternatively, at least one of the size, position, and brightness of the resin piece SP may be detected by the detection unit 20. In this case, a target resin piece SPt to be subjected to measurement processing by the measurement unit 30 can be selected based on only the items detected by the detection unit 20 out of the size, position, and brightness of the resin piece SP.

[0041] Steps S203 to S209 are processes performed by the control unit 40. Hereinafter, the resin piece SP identified in step S201 and whose size, position, and brightness have been calculated in step S202 may be referred to simply as the "resin piece SP."

[0042] In step S203, the control unit 40 determines the value of each resin piece SP. The value of each resin piece SP represents the degree to which it can contribute to improving the recovery rate of a specific type of resin piece SP, and in this embodiment, this may be the size of each resin piece SP. Therefore, the control unit 40 may determine the value of each resin piece SP based on the size of each resin piece SP detected by the detection unit 20. As an example, if the size of a resin piece SP detected by the detection unit 20 is 50 mm, ２ , the control unit 40 can determine the value of the resin piece SP as "50." Here, if the weight of each resin piece SP can be estimated from the size of each resin piece SP, the weight of each resin piece SP may be determined as the value of each resin piece SP. In this case, the control unit 40 can estimate the weight of the resin piece SP based on the size of the resin piece SP detected by the detection unit 20, and determine the value of the resin piece SP based on the estimated weight.

[0043] In step S204, the control unit 40 determines the time (measurement time) required for the measurement process in the measurement unit 30 for each resin piece SP. For example, the measurement time varies depending on the reflectance (color) of the resin piece SP. In the measurement process, it is necessary to irradiate the resin piece SP with the irradiation light 31 until the amount of light reaches a level sufficient for analyzing the Raman spectrum (reflected light). Therefore, the lower the reflectance of the resin piece SP, the longer the measurement time tends to be. Furthermore, the reflectance (color) of the resin piece SP corresponds to the brightness of the resin piece SP. Therefore, the control unit 40 can determine the measurement time according to the brightness of the resin piece SP obtained by the detection unit 20.

[0044] As an example, the control unit 40 may acquire information indicating the relationship between brightness and measurement time in advance, and based on that information, determine the measurement time from the brightness of the resin piece SP obtained by the detection unit 20. Alternatively, the control unit 40 may determine the measurement time to be a first time (e.g., 1000 ms) when the brightness of the resin piece SP obtained by the detection unit 20 is less than a threshold value, and may determine the measurement time to be a second time (e.g., 200 ms) that is shorter than the first time when the brightness is equal to or greater than the threshold value.

[0045] In step S205, the control unit 40 determines the time interval (hereinafter, sometimes referred to as the entry interval) at which the multiple resin pieces SP being transported on the belt conveyor 10 enter the measurement area 30'. For example, the control unit 40 can determine the entry interval based on the position of each resin piece SP (specifically, the leading edge position of each resin piece SP in the transport direction) detected by the detection unit 20. The control unit 40 can determine the entry interval for each of the multiple resin pieces SP being transported on the belt conveyor 10, the leading edges of which are adjacent to each other in the transport direction.

[0046] As an example, as shown in FIG. 2B , assume that the multiple resin pieces SP being transported on the belt conveyor 10 include a first resin piece SP1 and a second resin piece SP2 whose leading edge in the transport direction passes through the detection area 20′ next after the first resin piece SP1. In this case, the control unit 40 can determine the difference D between the positions of the leading edges of the first resin piece SP1 and the second resin piece SP2 in the transport direction and divide this difference D by the transport speed of the belt conveyor 10 to determine the resulting value as the penetration interval. Alternatively, the control unit 40 may determine the penetration interval based on the timing at which the leading edges of the resin pieces SP in the transport direction are first imaged by the imaging unit 21. In this case, the control unit 40 can determine the penetration interval as the difference between the timing t1 at which the leading edge of the first resin piece SP1 is first imaged by the imaging unit 21 and the timing t2 at which the leading edge of the second resin piece SP is first imaged by the imaging unit 21.

[0047] Here, the entry interval determined in step S205 is used as an index showing the density of the multiple resin pieces SP being transported on the belt conveyor 10. Therefore, it is not limited to the entry interval, as long as it can serve as the index. For example, in step S205, the control unit 40 may determine the frequency of passage of the resin pieces SP through the detection area 20' or the area occupied by the resin pieces SP in the width direction as the index, instead of the entry interval, based on the detection result of the detection unit 20.

[0048] In step S206, the control unit 40 determines whether it is possible to perform measurement processing on all of the resin pieces SP being transported on the belt conveyor 10 based on the measurement time determined in step S204 and the penetration interval determined in step S205. In the measurement unit 30 of this embodiment, the scanning mechanism 32 (galvanometer scanner) irradiates each resin piece SP with the irradiation light 31, so measurement processing is performed serially on each resin piece SP. Furthermore, the measurement area 30' is limited by the range of motion of the scanning mechanism 32 (galvanometer scanner). Therefore, depending on the time required to perform measurement processing on each resin piece SP and the density of the multiple resin pieces SP, it may not be possible to perform measurement processing on all of the multiple resin pieces SP. Therefore, in step S206, the control unit 40 determines whether it is possible to perform measurement processing on all of the resin pieces SP based on the measurement time determined in step S204 and the penetration interval determined in step S205.

[0049] If it is determined that the measurement process can be performed on all the resin pieces SP, the process proceeds to step S207, where the control unit 40 selects all the resin pieces SP as target resin pieces SPt. On the other hand, if it is determined that the measurement process cannot be performed on some of the resin pieces SP, the process proceeds to step S208.

[0050] In step S208, the control unit 40 selects a target resin piece SPt from the plurality of resin pieces SP based on the value of each resin piece SP determined in step S203. In this embodiment, the control unit 40 selects a target resin piece SPt from the plurality of resin pieces SP so that the larger the value determined in step S203 (i.e., the size detected by the detection unit 20) is, the higher the measurement process will be.

[0051] As an example, as shown in FIG. 2B , assume that the multiple resin pieces SP transported on the belt conveyor 10 include a first resin piece SP1 (first object) and a second resin piece SP2 (second object). The second resin piece SP2 is larger than the first resin piece SP1 and enters the measurement area 30′ after the first resin piece SP1. In this example, the control unit 40 determines whether or not measurement processing of the first resin piece SP1 and the second resin piece SP2 can be performed based on the detection results of the detection unit 20. Then, if the control unit 40 determines that measurement processing of the first resin piece SP1 would prevent measurement processing of the second resin piece SP2, it selects the target resin piece SPt so that measurement processing of the second resin piece SP2 is prioritized over that of the first resin piece SP1. That is, the control unit 40 does not select the first resin piece SP1 as the target resin piece SPt, but selects the second resin piece SP2 as the target resin piece SPt.

[0052] In step S209, the control unit 40 causes the measurement unit 30 to perform measurement processing on the target resin piece SPt determined in step S206 or S207. For example, the control unit 40 controls the scanning mechanism 32 to scan the irradiation light 31 in accordance with the movement of the target resin piece SPt, based on the position of the target resin piece SPt detected by the detection unit 20, so that the irradiation light 31 is irradiated onto the target resin piece SPt for a predetermined time. The measurement time determined in step S204 may be used as the predetermined time.

[0053] Next, a specific example of a selection method for selecting a target resin piece SPt from among a plurality of resin pieces SP being transported on the belt conveyor 10 will be described. Here, an example of selecting a target resin piece SPt using a beam search method will be described. The selection of the target resin piece SPt can be performed by the control unit 40.

[0054] 5 shows an example of selecting a target resin piece SPt from the four resin pieces SPa to SPd by the beam search method based on the value determined for each of the four resin pieces SPa to SPd (i.e., the size of each resin piece). Note that while four resin pieces SPa to SPd are shown in FIG. 5, the number of resin pieces SP is not limited to four, and may be two to three, or five or more.

[0055] In the example of FIG. 5 , in step S203, the value of resin piece SPa is determined to be "1," the value of resin piece SPb to be "4," the value of resin piece SPc to be "32," and the value of resin piece SPd to be "4." In FIG. 5 , a branch is set for each resin piece SPa to SPd as to whether or not to select it as a target resin piece SPt ("Yes" or "No"), and the total value of the value obtained by proceeding along each path is represented as score S. Although not shown in FIG. 5 , the total value of the measurement time obtained by proceeding along each path can also be calculated. Here, it is assumed that resin pieces SPa to SPd enter the measurement area 30' in that order. Furthermore, due to the entry interval determined in step S205, it is assumed that the measurement unit 30 cannot perform measurement processing on three or more resin pieces SP consecutively.

[0056] Looking at each path up to the third resin piece SPc, the condition under which the score S, which is the total value, is greatest is when resin piece SPb and resin piece SPc are selected as the target resin piece SPt. In this case, if half of the total value of resin piece SPb and resin piece SPc, "36," is set as the cutoff threshold, paths with a score S smaller than the cutoff threshold can be excluded. Furthermore, a time constraint is set for the measurement process to be performed by the measurement unit 30 within the measurement area 30', and paths for which the total measurement time does not satisfy the time constraint can also be excluded. As a result, the number of paths up to the fourth resin piece SPd can be reduced. Note that the above method is not limited to this; a specified number of paths may be excluded in order of lowest score S.

[0057] Next, looking at each path up to the fourth resin piece SPd, the path with the highest score S is the path represented by the thick line. That is, the path with resin pieces SPa, SPc, and SPd as target resin pieces SPt and resin piece SPb not as target resin piece SPt has the highest score S, and the score S is "37." In this way, by using the beam search method that sets the cutoff threshold and time constraints, it is possible to easily select the target resin piece SPt with the highest score S.

[0058] In the beam search method, the number of branches for each resin piece SP is not limited to two ("yes" and "no"), but may be three or more. Furthermore, the beam search method is merely an example, and a method other than the beam search method may be used to select the target resin piece SPt, as long as it takes into account the value of each resin piece SP, and, if necessary, the approach interval and measurement time.

[0059] [Operation Flow of Identification Process and Sorting Process] The operation flow of the identification process and sorting process will be described below. Fig. 6 is a flowchart showing the operation flow of the identification process of this embodiment. The flowchart of Fig. 6 starts when the detection unit 20 and the measurement unit 30 are operating, and ends when step S111 of the flowchart of Fig. 3 is executed. Furthermore, the flowchart of Fig. 6 can be executed for each target resin piece SPt that has been subjected to the measurement process by the measurement unit 30.

[0060] Here, the identification process refers to a process of identifying a specific type of resin piece SP from among a plurality of resin pieces SP being transported on the belt conveyor 10 based on the measurement results of the measuring unit 30. The identification process may include steps S301 to S304 in the flowchart of Fig. 6. The sorting process refers to a process of sorting a specific type of resin piece SP by the sorting unit 50. The sorting process may include step S305 in the flowchart of Fig. 6.

[0061] In step S301, the control unit 40 determines whether or not there is a target resin piece SPt that has been measured by the measurement unit 30. If there is no target resin piece SPt, step S301 is repeated, and if there is a target resin piece SPt, the process proceeds to step S302. In step S302, the control unit 40 acquires the measurement results of the target resin piece SPt by the measurement unit 30. Next, in step S303, the control unit 40 identifies the type (material) of the target resin piece SPt based on the measurement data acquired in step S302.

[0062] In this embodiment, the measurement unit 30 uses the scanning mechanism 32 to irradiate the target resin piece SPt with irradiation light 31 (excitation light laser) and the spectroscope 33 to separate the reflected light from the target resin piece SPt, thereby generating measurement data (spectral waveform) indicating the wavelength distribution of the reflected light. The control unit 40 stores reference data (reference spectral waveform) indicating the reference wavelength distribution of the reflected light for each type (material) of resin piece SP, which has been previously acquired through experiments, simulations, etc. for each type of resin piece SP. The control unit 40 acquires the measurement data from the measurement unit 30 and compares the measurement data with each type of reference data to calculate a correlation value (e.g., degree of agreement). This allows the control unit 40 to determine the type of reference data for which the correlation value is equal to or greater than a predetermined value (e.g., the type of reference data with the highest correlation value) as the type (material) of the target resin piece SPt.

[0063] In step S304, the control unit 40 determines whether the target resin piece SPt is a specific type of resin piece SP based on the identification result in step S303. A specific type of resin piece SP refers to a resin piece SP of a type (material) that the user wants to be sorted, and can be set by the user via a user interface, for example. If the target resin piece SPt is a specific type of resin piece SP, the process proceeds to step S305, where the control unit 40 controls the sorting unit 50 to emit compressed air toward the target resin piece SPt, and places the target resin piece SPt in the recovery box 62. On the other hand, if the target resin piece SPt is not a specific type of resin piece SP, the sorting unit 50 does not emit compressed air, and the target resin piece SPt is allowed to freely fall and placed in the recovery box 61.

[0064] As described above, the sorting device 100 (identification device) of this embodiment selects target resin pieces SP to be measured from among the multiple resin pieces SP transported on the belt conveyor 10 so that measurement is prioritized for pieces with larger values ​​(sizes). This prevents relatively large resin pieces SP of a specific type from not being measured, thereby improving the recovery rate of the specific type of resin pieces SP. Note that, when the transport speed of the belt conveyor 10 is increased, the recovery rate of the resin pieces SP tends to decrease. Even in such cases, this embodiment can prevent relatively large resin pieces SP of a specific type from not being measured. In other words, this embodiment has the effect of preventing a significant decrease in the recovery rate of the resin pieces SP even when the flow speed of the resin pieces SP is increased.

[0065] Second Embodiment A second embodiment of the present disclosure will be described. In the beam search method used to select target resin pieces SPt in the first embodiment described above, new target resin pieces SPt may be selected, including resin pieces SP for which it has already been determined whether or not to select them as target resin pieces SPt. Therefore, depending on the interval between the entry of multiple resin pieces SP and the conveying speed, there is a concern that the time required for the control unit 40 to select the target resin pieces SPt may be insufficient. Therefore, in this embodiment, an example of selecting target resin pieces SP using a simple method that does not use the beam search method will be described. This embodiment basically follows on from the first embodiment, and can follow the first embodiment except for the matters mentioned below.

[0066] In this embodiment, a value threshold TH1 (size threshold) for determining whether or not to perform measurement processing by the measurement unit 30 is set for the value (size) of the resin piece SP. As a result, the control unit 40 does not select resin pieces SP whose value is less than the value threshold TH1 as target resin pieces SPt, and selects only resin pieces SP whose value is equal to or greater than the value threshold TH1 as target resin pieces SPt. As shown in FIG. 7 , the value threshold TH1 may be changed according to the intrusion interval so that it becomes higher the shorter the intrusion interval (i.e., the higher the density of multiple resin pieces SP transported on the belt conveyor 10). Since the shorter the intrusion interval, the greater the number of resin pieces SP that cannot be measured by the measurement unit 30. Therefore, it is effective to set the value threshold TH according to the intrusion interval so that resin pieces SP with larger values ​​(sizes) are preferentially selected as target resin pieces SP.

[0067] The operation flow of the selection process and measurement process in this embodiment will be described below. Fig. 8 is a flowchart showing the operation flow of the selection process and measurement process in this embodiment. The flowchart in Fig. 8 starts in a state where the detection unit 20 and the measurement unit 30 are operating, and ends when step S111 of the flowchart in Fig. 3 is executed. Furthermore, during execution of the flowchart in Fig. 8, images of the detection area 20' can be continuously taken by the imaging unit 21 of the detection unit 20.

[0068] Steps S401 and S402 are steps performed by the processing unit 22 of the detection unit 20. In step S401, the processing unit 22 determines whether or not a resin piece SP is present in the image obtained by the imaging unit 21. If no resin piece SP is present in the image, step S401 is repeatedly executed, and if it is determined that a resin piece SP is present in the image, the processing unit 22 proceeds to step S402. In step S402, the processing unit 22 calculates (determines) the size, position, and brightness of each resin piece SP identified in step S401. Note that steps S401 and S402 are similar to steps S201 and S202 in the flowchart of FIG. 4, and therefore detailed description thereof will be omitted here.

[0069] Steps S403 to S410 are processes performed by the control unit 40. In step S403, the control unit 40 determines the value of each resin piece SP. In step S404, the control unit 40 determines the time (measurement time) required for measurement processing in the measurement unit 30 for each resin piece SP. In step S405, the control unit 40 determines the entry interval for multiple resin pieces SP being transported on the belt conveyor 10. Note that steps S403 to S405 are similar to steps S203 to S205 in the flowchart of FIG. 4, and therefore detailed description thereof will be omitted here.

[0070] In step S406, the control unit 40 determines whether the intrusion interval is equal to or greater than the interval threshold TH2 (i.e., whether the density of the multiple resin pieces SP is greater than the threshold). The interval threshold TH2 is a threshold for determining whether the measurement unit 30 can perform measurement processing on all of the multiple resin pieces SP being transported on the belt conveyor 10, and can be set in advance through experiments, simulations, etc. For example, as shown in FIG. 7 , if the intrusion interval is equal to or greater than the interval threshold TH2 (to the left of the interval threshold TH2), measurement processing can be performed on all of the resin pieces SP. On the other hand, if the intrusion interval is less than the interval threshold TH2 (to the right of the interval threshold TH2), measurement processing cannot be performed on some of the resin pieces SP, and a value threshold TH1 for selecting target resin pieces SPt can be set according to the intrusion interval.

[0071] If the penetration interval is equal to or greater than the interval threshold TH2, the process proceeds to step S407, where the control unit 40 selects all of the resin pieces SP as target resin pieces SPt. On the other hand, if the penetration interval is less than the interval threshold TH2, the process proceeds to step S408, where the control unit 40 sets a value threshold TH1 according to the penetration interval determined in step S405. Next, in step S409, the control unit 40 selects each resin piece SP having a value (size) equal to or greater than the value threshold TH1 as a target resin piece SPt from among the multiple resin pieces SP.

[0072] In step S410, the control unit 40 causes the measurement unit 30 to perform measurement processing on the target resin piece SPt determined in step S407 or S409. Note that step S410 is the same process as step S209 in the flowchart of FIG. 4, and therefore detailed description thereof will be omitted here.

[0073] Here, the function representing the change in the value threshold TH1 relative to the intrusion interval is not limited to a linear function, and may be, for example, a sigmoid function, a polynomial function, a step function, etc. Figures 9A to 9C show examples of functions representing the change in the value threshold TH1 relative to the intrusion interval. Figure 9A shows the function representing the change in the value threshold TH1 relative to the intrusion interval using a sigmoid function, Figure 9B shows the function representing the change in the value threshold TH1 relative to the intrusion interval using a polynomial function, and Figure 9C shows the function representing the change in the value threshold TH1 relative to the intrusion interval using a step function.

[0074] Furthermore, the function representing the change in the value threshold TH1 with respect to the intrusion interval may be changed according to the measurement time determined based on the brightness (reflectance) of each resin piece SP. For example, consider a case in which the measurement time for resin pieces SP whose brightness is below the threshold is determined to be a first time (1000 ms), and the measurement time for resin pieces SP whose brightness is equal to or greater than the threshold is determined to be a second time (200 ms). In this case, if the measurement time is also considered to represent the value of the resin piece SP (i.e., the shorter the measurement time, the higher the value), then a resin piece SP whose measurement time is determined to be the second time is five times more valuable than a resin piece SP whose measurement time is determined to be the first time. Therefore, as shown in FIG. 10 , by changing the function representing the change in the value threshold TH1 with respect to the intrusion interval according to the measurement time, the number of resin pieces SP that can be measured by the measurement unit 30 can be increased, thereby improving the recovery rate of specific types of resin pieces SP. Figure 10 shows a value threshold TH11 applied to a resin piece SP whose measurement time has been determined to be the first hour, and a value threshold TH12 applied to a resin piece SP whose measurement time has been determined to be the second hour.

[0075] As described above, the sorting device 100 (identification device) of this embodiment selects target resin pieces SP for measurement processing from among multiple resin pieces SP transported on the belt conveyor 10 depending on whether their value (size) is equal to or greater than the value threshold TH1. This also makes it possible to avoid the situation where measurement processing is not performed on relatively large resin pieces SP of a specific type, thereby improving the recovery rate of the specific type of resin pieces SP. Note that, when the transport speed of the belt conveyor 10 is increased, the recovery rate of the resin pieces SP tends to decrease. Even in such cases, this embodiment makes it possible to avoid the situation where measurement processing is not performed on relatively large resin pieces SP of a specific type. In other words, this embodiment has the effect of suppressing a significant decrease in the recovery rate of the resin pieces SP even when the flow speed of the resin pieces SP is increased.

[0076] <Embodiment of Article Manufacturing Method> The sorting device (identification device) according to the above embodiment can be used in an article manufacturing method for manufacturing articles such as plastic products. This article manufacturing method includes, for example, a sorting step of sorting plastic pieces using the sorting device (identification device) and a manufacturing step of manufacturing articles by processing the plastic pieces sorted in the sorting step. The processing of the plastic pieces can include, for example, at least one of melting, molding, cutting, assembly, inspection, etc. The article manufacturing method of this embodiment is advantageous over conventional methods in terms of article performance, quality, productivity, production costs, etc.

[0077] The technology described in this specification can contribute to the realization of a sustainable society, such as a carbon-free / recycling-based society.

[0078] The technical ideas derived from this disclosure are not limited to the disclosed exemplary embodiments, but are intended to encompass various modifications to the exemplary embodiments, or the replacement of equivalent structures or functions, etc. The scope of the following claims should be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.

[0079] This application claims priority based on Japanese Patent Application No. 2024-126144, filed August 1, 2024, the entire contents of which are incorporated herein by reference.

[0080] 10: Belt conveyor, 20: Detection unit, 30: Measurement unit, 32: Scanning mechanism, 33: Spectrometer, 40: Control unit, 50: Sorting unit, 100: Classification device

Claims

1. An identification device that identifies a specific type of object from among a plurality of objects being transported on a transport path, comprising: a detection unit that detects the size of each of the plurality of objects, including a first object and a second object, in a detection area of ​​the transport path; a measurement unit that, in a measurement area on the transport path downstream of the detection area, irradiates light onto an object selected from the plurality of objects and measures the light reflected from the object; and a control unit that identifies the specific type of object based on the measurement results of the measurement unit, wherein, when the control unit determines that it is unable to measure both the first object and the second object, the control unit determines the object to be measured by the measurement unit based on the sizes of the first object and the second object detected by the detection unit.

2. The identification device described in claim 1, characterized in that the control unit selects an object to be measured by the measurement unit from among the plurality of objects detected by the detection unit so that the larger the size of the object, the higher the priority of measurement by the measurement unit.

3. The identification device described in claim 2, characterized in that the second object has a size larger than the first object and enters the measurement area after the first object, and when the control unit determines that measuring the first object means that the second object cannot be measured by the measurement unit, it determines the object to be measured by the measurement unit based on the size of the first object and the size of the second object detected by the detection unit so that the second object is measured by the measurement unit with priority over the first object.

4. The identification device described in claim 2 or 3, characterized in that the detection unit further detects the position of each of the plurality of objects in the conveying direction, and the control unit determines whether or not to select an object from the plurality of objects to be measured by the measurement unit based on the density of the plurality of objects determined from the position of each object detected by the detection unit.

5. An identification device as described in any one of claims 2 to 4, characterized in that the detection unit further detects the brightness of each of the plurality of objects, and the control unit determines whether or not to select an object from the plurality of objects to be measured by the measurement unit based on the measurement time of the measurement unit determined for each object from the brightness detected by the detection unit.

6. An identification device according to any one of claims 1 to 5, characterized in that the control unit calculates the weight of each of the plurality of objects based on the size detected by the detection unit, and selects from the plurality of objects a measurement target to be measured by the measurement unit so that the heavier the object, the higher the priority for measurement by the measurement unit.

7. An identification device as described in any one of claims 1 to 6, characterized in that the control unit selects, from among the plurality of objects, an object whose size detected by the detection unit is equal to or greater than a threshold value as the object to be measured by the measurement unit.

8. The identification device described in claim 7, characterized in that the detection unit further detects the position of each of the plurality of objects in the conveying direction, and the control unit changes the threshold value according to the density of the plurality of objects determined from the position of each object detected by the detection unit.

9. The identification device described in claim 7 or 8, characterized in that the detection unit further detects the brightness of each of the plurality of objects, and the control unit changes the threshold value in accordance with the measurement time of the measurement unit determined for each object from the brightness detected by the detection unit.

10. An identification device as described in any one of claims 1 to 9, characterized in that the measurement unit includes a scanning mechanism that scans the light in the measurement area and a spectroscope that disperses the reflected light, and the control unit controls the scanning of the light by the scanning mechanism to follow the movement of an object moving in the measurement area so that the light is irradiated onto the object for a predetermined period of time.

11. A sorting device for sorting a plurality of objects, comprising: a conveying mechanism for conveying the plurality of objects on a conveying path; an identification device according to any one of claims 1 to 10 for identifying a specific type of object from the plurality of objects being conveyed on the conveying path by the conveying mechanism; and a sorting mechanism for selecting the specific type of object from the plurality of objects conveyed on the conveying path based on the identification result of the identification device.

12. A method for manufacturing an article, comprising: a classification step of classifying a plurality of objects using the classification device according to claim 11; and a manufacturing step of manufacturing an article by processing the specific type of object obtained through the classification step.

13. An identification method for identifying a specific type of object from a plurality of objects being transported on a transport path, comprising: a detection step of detecting the size of each of a plurality of objects, including a first object and a second object, in a detection area of ​​the transport path; a measurement step of irradiating light onto an object selected from the plurality of objects in a measurement area of ​​the transport path downstream of the detection area and measuring the light reflected from the object; and an identification step of identifying the specific type of object based on the measurement results of the measurement step, wherein if it is determined that both the first object and the second object cannot be measured, the object to be measured in the measurement step is determined based on the size of the first object and the size of the second object detected in the detection step.

14. A classification method for classifying a plurality of objects, comprising: using the identification method described in claim 13 to identify a specific type of object from among the plurality of objects being transported on a transport path; and sorting the plurality of objects based on the identification results.

Images