Sorting device and sorting method

Abstract

To provide a sorting device capable of suppressing erroneous recognition and decrease in throughput.SOLUTION: There is provided a sorting device comprising: conveyance means for conveying a workpiece using a conveyance belt; position detection means for detecting the position of the workpiece by using a difference between an image including the workpiece on the conveyance belt and a background image other than the workpiece; determination means for determining the type of the workpiece at the position detected by the position detection means; and update means for updating the background image.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to a separation device, a separation method, and the like.

Background Art

[0002] Recycling that uses plastic products as materials for resin parts again is required. However, the plastics to be recycled are a mixture of multiple types, and specific plastics must be sorted from a mixture containing various color types. For this purpose, a separation device (resin sorting device) that separates specific plastics from a mixture containing various plastics has been developed.

[0003] For example, as a resin sorting device, there is a device that irradiates resin with measurement light and analyzes the reflected light to identify the resin type. In that case, those using near-infrared spectroscopy are mainstream, but since carbon black used for coloring black plastics absorbs near-infrared light, it cannot be measured. On the other hand, in resin identification by Raman spectroscopy, it is possible to measure black plastics, so plastics can be reused to the maximum extent.

[0004] Patent Document 1 describes a method of identifying and separating resin types by obtaining Raman scattering signals from resin flowing on a belt conveyor.

[0005] Patent Document 2 discloses a method of irradiating Raman light following resin flowing on a belt conveyor to identify a large amount of resin at high speed. Measuring black plastics requires a longer measurement time than other colored plastics. By following the measurement light with respect to the moving sample, the measurement time is ensured.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0007] The sorting device of Patent Document 2 sets a path (hereinafter referred to as a tracking path) for irradiating a workpiece with measurement light. In order to set the tracking path based on the workpiece on the conveyor belt, it is necessary to recognize the workpiece before setting, and the workpiece is recognized from the difference in luminance between the conveyor belt and the workpiece. However, since there is also a difference in luminance with respect to dirt and scratches on the conveyor belt, there may be misrecognition as the workpiece.

[0008] When misrecognition occurs, a tracking path is set even for dirt and scratches, resulting in wasteful processing. The occurrence of wasteful processing due to the upper limit of the number of trackings per unit time may prevent the setting of a tracking path including the workpiece to be identified.

[0009] By reducing the number of workpieces to be conveyed or by reducing the conveyance speed, the number of workpieces that can be tracked can be increased, but the number of processes per unit time (hereinafter referred to as "throughput") decreases. In order to correctly recognize the workpiece and improve the efficiency of tracking, it is necessary to suppress misrecognition due to dirt and scratches on the conveyor belt and prevent a decrease in throughput.

[0010] One object of the present invention is to provide a sorting device that solves the above problems, suppresses misrecognition, and suppresses a decrease in throughput.

Means for Solving the Problems

[0011] To solve the above problems, the sorting device of the present invention a conveying means for conveying a workpiece using a conveyor belt; a position detecting means for detecting the position of the workpiece using the difference between an image including the workpiece on the conveyor belt and a background image other than the workpiece; Discrimination means for discriminating the type of the workpiece at the position detected by the position detection means; Updating means for updating the background image, and is characterized by having the same.

Advantages of the Invention

[0012] According to the present invention, it is possible to realize a sorting device that can suppress misrecognition and suppress a decrease in throughput.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Embodiments for Carrying Out the Invention

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the present invention is not limited to the following embodiments. In each figure, the same members or elements are denoted by the same reference numerals, and duplicate explanations are omitted or simplified. <Embodiment 1>

[0015] FIG. 1 is a diagram showing an example of the hardware configuration of the sorting device according to Embodiment 1 of the present invention. The sorting device 100 includes a CPU 101, an input / output I / F 102 to an external storage device, a ROM 104, and a RAM 105, and these perform data exchange via a bus 103.

[0016] In addition, a touch panel 106 and a control unit 107 are connected to the sorting device 100 and perform data exchange via the bus 103. Note that the touch panel 106 and the control unit 107 may be arranged in the housing of the sorting device, or may be provided in a housing separate from the sorting device.

[0017] FIG. 2 is a functional block diagram showing an example of the functional configuration of the sorting device according to Embodiment 1 of the present invention. Note that some of the functional blocks shown in FIG. 2 are realized by causing a CPU 101 as a computer included in the sorting device to execute a computer program stored in a memory as a storage medium.

[0018] However, some or all of them may be realized by hardware. As the hardware, a dedicated circuit (ASIC), a processor (reconfigurable processor, DSP), or the like can be used. In addition, each of the functional blocks shown in FIG. 2 does not have to be built in the same housing, and may be configured by separate devices connected to each other via a signal path.

[0019] The control unit 107 also incorporates a CPU or the like as a computer, and functions as control means for controlling the operations of each part of the sorting device and the entire sorting system based on a computer program stored in a memory as a storage medium.

[0020] For example, external units such as an irradiation unit for near-infrared light or laser light, a camera, and an air jet included in the sorting system are controlled by the control unit 107. Also, the touch panel 106 is used for starting / ending instructions for the sorting device 100 and outputting the current state.

[0021] The sorting device operates as a device that sequentially identifies workpieces on the workpiece conveyance unit and separates them according to the identification results. In this embodiment, the object conveyed as the workpiece is not limited, but here an example where, for example, resin is conveyed as the workpiece will be described.

[0022] The sorting device 100 has a workpiece conveyance unit 201. The workpiece conveyance unit is, for example, a belt conveyor or the like that carries resin. Here, the workpiece conveyance unit 201 functions as conveyance means for executing a conveyance step of conveying the workpiece using a conveyance belt.

[0023] The workpiece position, which is the position of a workpiece such as resin conveyed by the workpiece conveyance unit, is detected by the workpiece position detection unit 202. At that time, for example, the workpiece position is detected by the difference between a background image of the belt conveyance surface without the workpiece mounted and an image including the workpiece on the conveyance belt. Here, the workpiece position detection unit 202 functions as position detection means for detecting the position of the workpiece using the difference between an image including the workpiece on the conveyance belt and a background image other than the workpiece.

[0024] For the workpiece at the detected workpiece position, the workpiece identification unit 203 identifies the type of resin. For example, for the workpiece such as the resin, laser irradiation is sequentially performed along a predetermined tracking path to obtain a Raman scattering signal, thereby identifying the type of resin.

[0025] Alternatively, the type of the resin may be identified by irradiating near-infrared light onto the resin and analyzing the reflected and transmitted light. In this embodiment, these identification means are not limited. Here, the work identification unit 203 functions as a discrimination means for discriminating the type of the work at the position detected by the position detection means.

[0026] When the identification process is completed in the work identification unit 203, upon receiving the result of the work identification unit, the work separation unit 204 separates the resin to be identified from the resins other than the target. For example, the resins are separated and stored in a box where the resin to be identified should be stored according to its type.

[0027] Each of these functional units is realized by the CPU of the control unit 107 expanding the computer program stored in the ROM into the RAM and executing the processes according to the respective flowcharts described later. Also, for example, when configuring hardware as at least a part of the software process using the above CPU, an arithmetic unit or a circuit corresponding to the processes of each functional unit described here may be configured.

[0028] Next, the configuration of an apparatus (hereinafter referred to as a separation system) that sequentially identifies works such as resins supplied onto a belt conveyor incorporating the apparatus of this embodiment and separates them for each identification result will be described.

[0029] FIG. 3 is a view of a hardware configuration example of a separation system 300 according to Embodiment 1 of the present invention as seen from the side. The separation system 300 supplies a work 303 such as a resin supplied from an external belt conveyor 301 to the belt conveyor 302 of the separation system 300 to perform resin separation.

[0030] The separation system 300 is equipped with a work conveyance surface imaging camera 304, a conveyance belt imaging camera 305, a length measuring instrument 306 for measuring the moving amount of the conveyance belt, a work identification camera 307, and a laser irradiation device 308.

[0031] The work transfer surface imaging camera 304 is used by the work position detection unit 202 to detect the position of a work 303 such as resin conveyed by a belt conveyor. The conveyor belt imaging camera 305 is used to acquire an image of the belt conveyance surface without the work 303 mounted thereon as a background image.

[0032] The length measuring instrument 306 is used to detect the amount of movement and the moving speed of the belt. Also, the work identification camera 307 and the laser irradiation device 308 are used by the work identification unit 203 to identify the type of resin.

[0033] The air jet 309 discharges air toward the resin according to the operation of the work separation unit 204 of the device of the present embodiment. Thereby, the work 303 such as resin to be recovered is controlled to enter the recovery box 310, and the resin to be discarded enters another box 311.

[0034] FIG. 4 is a flowchart showing a processing example of a separation system using the separation device according to Embodiment 1 of the present invention, and FIG. 5 is a flowchart showing an operation flow example of the work separation process (step S404) in FIG. 4. Note that the operations of the respective steps of the flowcharts in FIGS. 4 and 5 are sequentially performed by the CPU or the like as a computer in the separation device 100 or the control unit 107 executing a computer program stored in a memory.

[0035] Using FIGS. 4 and 5, the flow of the separation operation of the work 303 such as resin will be described. First, when an operation start is instructed on the touch panel 106, the flow in FIG. 4 starts. In step S401, the CPU performs a preparation process. That is, the supply of power to each unit (air jet, camera, laser irradiation device, belt conveyor, etc.) constituting the separation system 300 is started, and these units are shifted to an operable state.

[0036] When the preparation operation is completed, at step S402, the CPU determines whether a stop signal has been received. The stop signal may be issued when an instruction to stop the operation is given on the touch panel 106, or may be issued when a safety problem occurs in the system.

[0037] If it is determined at step S402 that the stop signal has been received, the process proceeds to step S405, where end processing is performed, the system is immediately stopped, and the flow of FIG. 4 ends.

[0038] If it is not determined at step S402 that the stop signal has been received, at step S403, the CPU determines whether the workpiece 303 exists on the belt conveyor.

[0039] If it is determined at step S403 that the workpiece 303 does not exist, the process returns to step S401. If it is determined at step S403 that the workpiece exists, the process proceeds to step S404, where the workpiece sorting process of the sorting device is performed. The details of step S404 will be described with reference to FIG. 5.

[0040] FIG. 5 is a flowchart showing an example of the operation flow of the workpiece sorting process (step S404) in FIG. 4. When entering the flow of FIG. 5, at step S501, a background image is acquired. That is, the background image is updated by periodically capturing an image of the conveyor belt immediately before the workpiece 303 is mounted by the conveyor belt imaging camera 305. Here, the conveyor belt imaging camera 305 functions as an imaging means for imaging the conveying surface of the conveyor belt.

[0041] Here, step S501 functions as an update step (update means) for updating the background image, and updates the background image based on the image acquired by the imaging means when no workpiece is mounted on the conveying surface of the conveyor belt. Note that dirt on the conveyor belt includes, for example, partial changes in color and reflectivity due to the aging of the surface of the conveyor belt. That is, the background image includes the dirt or scratches on the conveyor belt.

[0042] After acquiring the background image, in step S502, an image of the work mounting surface on which the work 303 is mounted is acquired. That is, in order to acquire an image of the work mounting surface, an image of the work 303 on the conveyor belt is captured by the work conveyor surface imaging camera 304.

[0043] After acquiring the image of the work mounting surface, in step S503, differential processing is performed to recognize the work 303 by threshold processing on the luminance values obtained from the difference between the images of the conveyor belt and the work mounting surface. In addition, in order to make the luminance values of the work conveyor surface imaging camera 304 and the conveyor belt imaging camera 305 the same, the same lighting system is used, or the luminance levels and gain of the work conveyor surface imaging camera 304 and the conveyor belt imaging camera 305 are adjusted.

[0044] Thus, in this embodiment, in step S501, a background image immediately before the work 303 is mounted is acquired, and background image update processing is performed in step S503 using the background image. In addition, in this embodiment, within one rotation of the conveyor belt, the background image including dirt or scratches is updated at least once.

[0045] Next, in step S504, the CPU performs work position detection processing. That is, the size and position of the work 303 are detected by comparing the luminance data obtained by the differential processing with a predetermined threshold value. Here, step S504 functions as a position detection step for detecting the position of the work using the difference between the image including the work on the conveyor belt and the background image other than the work.

[0046] Next, in step S505, the CPU performs setting of the tracking path of the laser beam by the laser irradiation device 308 based on the recognized work 303. Here, step S505 functions as a tracking path setting step (tracking path setting means) for setting the tracking path based on the position of the work.

[0047] Figs. 6(A) and 6(B) are diagrams showing examples of setting a tracking path according to Embodiment 1 of the present invention, and Fig. 6(A) is a diagram showing an example of a tracking path. A tracking path 603 is set, which is a path for sequentially irradiating laser light based on the positions of the recognized workpieces 303 in order from the head in the conveyance direction.

[0048] As shown in (2) and (2)' of Fig. 6(A), when the recognized workpieces are arranged perpendicular to the conveyance direction, for example, the workpiece on the upper side in Fig. 6(A) is preferentially determined for tracking. At this time, if the dirt or scratches 602 on the conveyance belt are misrecognized as workpieces, a tracking path including the dirt or scratches will be set.

[0049] Fig. 6(B) is a diagram showing an example of setting a tracking path when position detection is performed using the background image immediately before including dirt or scratches in the present embodiment. By performing differential processing with the background image immediately before including the dirt or scratches 602, a path for tracking only the workpiece 303 can be generated.

[0050] As a result, even if new dirt or scratches 602 occur, they will not be misrecognized as workpieces, and the workpieces can be efficiently tracked. Further, after setting the tracking path in step S505, in step S506, the CPU performs identification processing of the resin type of the workpiece 303.

[0051] For this purpose, the laser irradiation device 308 irradiates laser light to each workpiece 303 based on the set tracking path, and for example, the resin type is identified by imaging the reflected light (Raman scattered light) with the workpiece identification camera 307.

[0052] Here, step S506 functions as a discrimination step for discriminating the type of the workpiece at the position detected by the position detection step. Regarding the identification of the resin using Raman scattered light, a known technique may be used, and the description is omitted here.

[0053] According to the identification result by the identification process in step S506, in step S507, the CPU performs workpiece separation. For example, according to the identified resin type, the air jet 309 is controlled so that the workpiece 303 is sorted and stored in the collection box 310 or another box 311. When these series of processes are completed, the flow in FIG. 5 ends.

[0054] As described above, in this embodiment, immediately before performing the workpiece identification process, in order to update the background image including locations with different brightness from the conveyor belt, such as newly generated dirt and scratches on the conveyor belt, the identification process for dirt and scratches on the conveyor belt can be reduced. Therefore, it is possible to prevent a decrease in throughput.

[0055] In addition, when separating the workpiece 303, an example of separating it into two boxes, the collection box and another box, is shown. However, this embodiment is not limited to the example of separating it into two boxes. It may be separated into three or more boxes for each type of resin such as ABS, polystyrene, polyethylene, and polypropylene.

[0056] <Embodiment 2> FIG. 7 is a view of the hardware configuration of the sorting system according to Embodiment 2 as seen from the side. For the same configuration as in FIG. 5, the same numbers are assigned. The example in FIG. 5 is different in that there is no camera for acquiring the background image.

[0057] FIG. 8 is a flowchart showing an example of the workpiece sorting operation flow according to Embodiment 2. Note that the operations of each step in the flowchart of FIG. 8 are sequentially performed by the CPU or the like as a computer in the sorting device 100 or the control unit 107 executing a computer program stored in the memory.

[0058] When the flow in FIG. 8 starts, the CPU acquires an image on which the workpiece 303 is mounted in step S801, and performs a difference process between the background image acquired in advance and the image of the workpiece when the workpiece 303 is mounted on the conveyor belt in step S802. Further, the CPU performs a workpiece position detection process based on the data of the difference process in step S803.

[0059] Next, in step S804, the CPU sets a tracking path for the recognized workpiece, and in step S805, performs workpiece identification processing using the Raman scattered light of the laser beam.

[0060] Also, in step S806, the CPU performs workpiece separation according to the identification result from the identification processing. That is, according to the identified resin type, the air jet 309 is controlled so that the workpiece 303 such as resin is stored in the collection box 310 or another box 311.

[0061] Next, in step S807, the CPU determines whether new dirt or scratches have occurred on the conveyor belt based on the identification result from the workpiece identification processing. As shown in Fig. 6(A), even if dirt or scratches are misjudged as workpieces due to the difference in luminance from the conveyor belt, since the Raman scattering signals are different between the conveyor belt and the workpiece, it is possible to determine the detected dirt or scratches.

[0062] When it is determined in step S807 that dirt or scratches are detected, in step S808, the CPU performs background image update processing including the detected dirt or scratches in the background.

[0063] Here, step S808 functions as an update step (update means) for determining dirt or scratches based on the Raman scattering signal from the dirt or scratches on the conveyor belt and updating the background image based on the determined dirt or scratches. Note that the update means updates the background image periodically. After the processing of step S808, or when it is determined as No in step S807, the series of flows from step S801 to step S808 in Fig. 8 is terminated.

[0064] As described above, in the second embodiment, based on the result of the workpiece identification processing, by detecting dirt or scratches, the background image including portions with different luminance from the conveyor belt, such as newly generated dirt or scratches on the conveyor belt, is updated. Therefore, in the next cycle, it becomes possible to prevent a decrease in throughput by not performing the identification processing on the dirt or scratches on the conveyor belt.

[0065] <Embodiment 3>

[0066] FIG. 9 is a flowchart showing an example of the work sorting operation of the sorting device according to Embodiment 3. Note that the operations of each step of the flowchart in FIG. 9 are sequentially performed by the CPU or the like as a computer in the sorting device 100 or the control unit 107 executing a computer program stored in the memory.

[0067] When the flowchart in FIG. 9 starts, in step S901, the CPU acquires an image on which a work is mounted, and in step S902, performs a difference process between the background image acquired in advance and the image on which the work is mounted. Also, in step S903, the CPU performs a work position detection process based on the result of the difference process.

[0068] After the work position detection process, in step S904, the CPU stores the positions of the detected works 303 respectively.

[0069] FIG. 10 is a schematic diagram of the area of the conveyor belt of the sorting device according to Embodiment 3. As shown in FIG. 10, the area of the conveyor belt is set by dividing the width direction of the conveyor belt into M pixels and the length direction of the belt into N pixels.

[0070] In step S904, the pixels that recognize the work 303 are stored, and the number of times 1001 that the work appears at those pixels is stored. Next, in step S905, the CPU sets a tracking path for the detected work 303, and in step S906, performs a work identification process. Also, in step S907, the CPU performs work separation according to the identification result by the identification process. For example, the air jet 309 is controlled so that the work 303 is stored in the collection box 310 or another box 311 according to the resin type of the identified work 303.

[0071] Furthermore, in step S908, the CPU determines whether the number of times a work appears in each pixel shown in FIG. 10 exceeds a predetermined threshold. Dirt and scratches that cause misrecognition on the conveyor belt always appear as areas with different luminance from the conveyor belt.

[0072] Therefore, when it is determined Yes in step S908, in step S909, the CPU determines that the pixels where the number of times the work appears exceeds the predetermined threshold are dirt or scratches, and registers them in the background image.

[0073] That is, the number of appearances of areas with different luminance from the conveyor belt is added up. When the number of appearances exceeds the predetermined threshold, it is determined Yes in step S908, and the process proceeds to step S909. Then, in step S909, the CPU performs background image update processing to generate a background image including the area 1002 that exceeds the threshold (for example, 2).

[0074] Here, step S909 functions as an update step (update means) for updating the background image by regarding the areas where the number of appearances as the position of the work is equal to or more than the predetermined threshold as dirt or scratches. On the other hand, when the number of appearances does not exceed the threshold, that is, when it is determined No in step S908, a series of processes are completed, and the flow of FIG. 9 ends.

[0075] As described above, in Embodiment 3, by detecting dirt and scratches based on the number of appearances of the areas recognized as the work, the background image including the areas with different luminance from the conveyor belt such as newly generated dirt and scratches on the conveyor belt is updated. Therefore, it becomes possible to prevent a decrease in throughput without performing discrimination processing on the dirt and scratches on the conveyor belt.

[0076] <Embodiment 4> FIG. 11 is a side view of the hardware configuration of the sorting system according to Embodiment 4. Note that the components having the same configuration as those in FIG. 5 are given the same numbers. The difference from the configuration of FIG. 5 is that a 3D camera 1101 is used as the work position detection means.

[0077] FIG. 12 is a flowchart showing an example of a workpiece sorting operation flow of the sorting apparatus according to Embodiment 4. Note that, by the CPU or the like as a computer in the sorting apparatus 100 or the control unit 107 executing a computer program stored in a memory, the operations of each step of the flowchart in FIG. 12 are sequentially performed.

[0078] In step S1201, the CPU acquires three-dimensional data of the workpiece. That is, by using the three-dimensional camera 1101, in addition to the position and size of the workpiece, height information is acquired.

[0079] Next, in step S1202, the CPU performs a workpiece position detection process. At this time, the height of the workpiece is compared with a predetermined threshold value, and those equal to or greater than the predetermined threshold value are recognized as workpieces, and the workpiece position is detected for those recognized as workpieces. Here, step S1202 functions as a position detection step for detecting the workpiece position by discriminating the height of the workpiece based on a predetermined threshold value.

[0080] On the other hand, those less than the predetermined threshold value may be registered in the background image as, for example, dirt or scratches, and the background image may be updated. That is, step S1202 can function as an update step (update means) for updating the background image by registering an area where the height of the object on the conveyor belt is less than a predetermined threshold value as dirt or scratches.

[0081] Next, in step S1203, the CPU sets a tracking path based on the detected workpiece position. Further, in step S1204, the CPU irradiates the workpiece on the tracking path with laser light and identifies the resin type of the workpiece based on the Raman scattered light. Then, in step S1205, the CPU separates the workpiece according to the identified resin type.

[0082] In this way, a workpiece with a height equal to or higher than a predetermined threshold is recognized as a workpiece such as resin, and a workpiece with a height lower than the predetermined threshold is recognized as dirt or scratches on the conveyor belt, so that dirt or scratches on the conveyor belt are not misrecognized as workpieces. Therefore, a correct tracking path can be set, and misrecognition of workpieces can be reduced.

[0083] In this way, in Embodiment 4, since the workpiece is measured three-dimensionally, it is possible to prevent incorrect discrimination processing from being performed on dirt or scratches on the conveyor belt, and it is possible to prevent a decrease in throughput.

[0084] In addition, in Embodiment 4, the height of the workpiece is obtained by a three-dimensional camera, but any sensor may be used as long as it can measure and obtain the height of the workpiece. For example, a distance sensor or a sensor such as LiDAR (Light Detection And Ranging) may be used to obtain the height of the workpiece.

[0085] <Embodiment 5> Next, a method for manufacturing an article (such as a semiconductor IC element, a liquid crystal display element, MEMS, etc.) using a sorting device or a sorting system such as those in Embodiments 1 to 4 will be described. For example, a substrate or the like is sorted using the sorting device or the sorting system according to the above embodiments. Then, for the substrate or the like, a step of applying a photosensitive material, a step of exposing a substrate (wafer, glass substrate, etc.) coated with a photosensitizer, a step of developing the substrate (photosensitizer), and a step of processing the developed substrate by other well-known steps are performed to manufacture an article.

[0086] In addition, other well-known steps include etching, resist stripping, dicing, bonding, packaging, etc. According to such an article manufacturing method, it is possible to manufacture high-quality articles without causing a decrease in throughput as compared with the prior art.

[0087] As described above, the present invention has been described in detail based on its preferred embodiments. However, the present invention is not limited to the above embodiments, and various modifications and combinations of the above embodiments are possible based on the spirit of the present invention, and they are not excluded from the scope of the present invention. The present invention includes the following combinations.

[0088] (Configuration 1) A sorting device comprising: a conveying means for conveying a workpiece using a conveyor belt; a position detecting means for detecting the position of the workpiece using the difference between an image including the workpiece on the conveyor belt and a background image other than the workpiece; a discriminating means for discriminating the type of the workpiece at the position detected by the position detecting means; and an updating means for updating the background image.

[0089] (Configuration 2) The sorting device according to Configuration 1, further comprising a tracking path setting means for setting a tracking path based on the position of the workpiece.

[0090] (Configuration 3) The sorting device according to Configuration 1 or 2, including an imaging means for imaging the conveying surface of the conveyor belt, wherein the updating means updates the background image based on an image acquired by the imaging means when the workpiece is not mounted on the conveying surface of the conveyor belt.

[0091] (Configuration 4) The sorting device according to any one of Configurations 1 to 3, wherein the updating means determines the dirt or damage based on a Raman scattering signal from the dirt or damage of the conveyor belt, and periodically updates the background image based on the determined dirt or damage.

[0092] (Configuration 5) The sorting device according to any one of Configurations 1 to 4, wherein the updating means updates the background image including the dirt or damage at least once within one rotation of the conveyor belt.

[0093] (Configuration 6) The sorting device according to any one of Configurations 1 to 5, wherein the updating means updates the background image using, as the dirt or damage, a region where the number of appearances as the position of the workpiece is equal to or greater than a predetermined threshold.

[0094] (Configuration 7) The separation device according to any one of Configurations 1 to 6, wherein the updating means updates the background image using, as the stain or scratch, a region where the height of an object on the conveyor belt is less than a predetermined threshold value.

[0095] (Configuration 8) The separation device according to any one of Configurations 1 to 7, wherein the background image includes stains or scratches on the conveyor belt.

[0096] (Configuration 9) The separation device according to any one of Configurations 1 to 8, wherein the updating means periodically updates the background image.

[0097] (Configuration 10) A separation device comprising: a conveyance means for conveying a workpiece; a position detection means for detecting the position of the workpiece; and a discrimination means for discriminating the type of the workpiece at the position of the workpiece detected by the position detection means, wherein the position detection means detects the position of the workpiece by discriminating the height of the workpiece based on a predetermined threshold value.

[0098] (Method 1) A separation method comprising: a conveyance step of conveying a workpiece using a conveyor belt; a position detection step of detecting the position of the workpiece using a difference between an image including the workpiece on the conveyor belt and a background image other than the workpiece; a discrimination step of discriminating the type of the workpiece at the position detected by the position detection step; and an updating step of updating the background image.

[0099] (Method 2) A separation method comprising: a conveyance step of conveying a workpiece; a position detection step of detecting the position of the workpiece; and a discrimination step of discriminating the type of the workpiece at the position of the workpiece detected by the position detection step, wherein the position detection step detects the position of the workpiece by discriminating the height of the workpiece based on a predetermined threshold value.

[0100] Furthermore, in order to implement some or all of the control in the above-described embodiment, a computer program for implementing the functions of the above-described embodiment may be supplied to a sorting device or the like via a network or various storage media. Then, a computer (or a CPU, MPU, etc.) in the sorting device or the like may read and execute the program. In that case, the program and the storage medium storing the program will constitute the present invention.

Explanation of Signs

[0101] 100: Sorting device 107: Control unit 201: Work transfer unit 202: Work position detection unit 203: Work identification unit 204: Work separation unit 300: Sorting system

Claims

1. Conveying means for conveying a workpiece using a conveyor belt, Position detection means for detecting the position of the workpiece using the difference between an image including the workpiece on the conveyor belt and a background image other than the workpiece, Discrimination means for discriminating the type of the workpiece at the position detected by the position detection means, Updating means for updating the background image, wherein the sorting device is characterized by comprising the above.

2. The sorting device according to claim 1, further comprising tracking path setting means for setting a tracking path based on the position of the workpiece.

3. Including imaging means for imaging the conveying surface of the conveyor belt, The updating means updates the background image based on an image acquired by the imaging means when the workpiece is not mounted on the conveying surface of the conveyor belt. The sorting device according to claim 1 is characterized by this.

4. The updating means determines the dirt or damage by Raman scattering signals from the dirt or damage on the conveyor belt, and periodically updates the background image based on the determined dirt or damage. The sorting device according to claim 1 is characterized by this.

5. The updating means updates the background image including the dirt or damage at least once within one rotation of the conveyor belt. The sorting device according to claim 1 is characterized by this.

6. The updating means updates the background image with a region where the number of appearances as the position of the workpiece is equal to or more than a predetermined threshold as the dirt or damage. The sorting device according to claim 1 is characterized by this.

7. The updating means updates the background image with a region where the height of an object on the conveyor belt is less than a predetermined threshold as the dirt or damage. The sorting device according to claim 1 is characterized by this.

8. The background image includes dirt or damage on the conveyor belt. The sorting device according to claim 1 is characterized by this.

9. The updating means periodically updates the background image. The sorting device according to claim 1 is characterized by this.

10. Conveying means for conveying a workpiece, Position detection means for detecting the position of the workpiece, Discrimination means for discriminating the type of the workpiece at the position of the workpiece detected by the position detection means, and The position detection means detects the position of the workpiece by discriminating the height of the workpiece based on a predetermined threshold. The sorting device is characterized by this.

11. A conveying step of conveying a workpiece using a conveyor belt, A position detection step of detecting the position of the work by using the difference between an image including the work on the conveying belt and a background image other than the work; A discrimination step of discriminating the type of the work at the position detected by the position detection step; An update step of updating the background image, wherein the sorting method is characterized by comprising these steps.

12. A conveying step of conveying the work; A position detection step of detecting the position of the work; A discrimination step of discriminating the type of the work at the position of the work detected by the position detection step by the position detection step; and The position detection step is characterized in that the position of the work is detected by discriminating the height of the work based on a predetermined threshold value.

Images