Battery sorting system

Abstract

To easily sort a battery or a waste product including the battery from a waste group.SOLUTION: The present battery sorting system includes (A) first and second conveyance units configured to convey waste including batteries, and (B) a control unit configured to, based on a first transmitted X-ray image of the waste being conveyed by the first conveyance unit, A first detecting unit for detecting a battery or a discarded article containing the battery contained in the waste material; (C) a transferring unit for transferring a part of the waste material containing the battery or the discarded article containing the battery to a second conveying unit when the battery or the discarded article containing the battery is detected by the first detecting unit; And a second detection part for detecting the battery included in a part of the waste or the waste product including the battery, and for outputting information for sorting the detected battery or waste product including the battery.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a system for sorting batteries contained in waste products and waste materials. [Background technology]

[0002] At recycling plants, there are as many as 9,000 fires per year caused by batteries contained in waste products, such as home appliances, or batteries mixed in with discarded waste containing flammable materials, non-flammable materials, plastics, etc. Many of these fires are caused by batteries being mistakenly thrown into a crusher. Therefore, there is a need for technology to automatically detect batteries in advance. In this application, primary batteries and secondary batteries are referred to as batteries. Furthermore, waste may include waste products.

[0003] For example, Non-Patent Document 1 discloses a technology for detecting batteries in waste products by photographing them with transmission X-rays and using a deep learning program. This Non-Patent Document 1 uses a single deep learning program that has been trained on all training data.

[0004] On the other hand, Non-Patent Document 2 discloses the following technology for detecting batteries while achieving both a high precision rate (the rate at which those detected by the artificial intelligence are truly batteries) and a high recall rate (the rate at which the artificial intelligence detects batteries among all batteries). Specifically, in an input transmission X-ray image of at least one of a discarded product and waste, an image portion in which one of a predetermined number of types of discarded products appears is detected, and a first battery detection unit, which corresponds to the type of discarded product appearing in the detected image portion and has been trained in advance to detect batteries built into the discarded product from the transmission X-ray image of that type of discarded product, is made to detect the battery appearing in the detected image portion. Furthermore, a second battery detection unit, which has been trained in advance to detect batteries from transmission X-ray images of at least the waste or the battery itself, is made to detect batteries appearing in an area of ​​the input transmission X-ray image other than the detected image portion.

[0005] While this technology for detecting batteries using transmitted X-ray images can determine the position of the battery within the plane of the transmitted X-ray image, it is not possible to determine the vertical position of the waste when the waste is piled up, which poses the problem that it is not easy to automatically separate batteries or waste products containing batteries from a pile of waste. For example, batteries buried in waste cannot be seen or approached from above, so robotic hands cannot be used, and if the waste is layered, it cannot be flicked with a paddle. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] W. Sterkens, et. al., "Detection and recognition of batteries on X-Ray images of waste electrical and electronic equipment using deep learning", Resources, Conservation and Recycling, Volume 168, May 2021, 105246 [Non-patent document 2] "Automatically detects whether batteries are present in waste without relying on worker experience"<https: / / www.aist.go.jp / aist_j / press_release / pr2023 / pr20230522 / pr20230522.html> Summary of the Invention [Problem to be solved by the invention]

[0007] Therefore, according to one aspect, it is an object of the present invention to provide a technique for easily separating batteries or waste products containing batteries from a waste mass. [Means for solving the problem]

[0008] A battery sorting system according to a first aspect of the present invention comprises: (A) first and second conveying units that transport waste including batteries; (B) a first detection unit that detects batteries or waste products including batteries contained in the waste based on a first transmitted X-ray image of the waste being transported by the first conveying unit; (C) a transfer unit that, when a battery or a waste product including a battery is detected by the first detection unit, transfers a portion of the waste including the battery or the waste product including a battery to a second conveying unit; and (D) a second detection unit that detects batteries or waste products including batteries contained in the portion of the waste based on a second image of the portion of the waste being transported by the second conveying unit, and outputs information for sorting the detected batteries or waste products including batteries.

[0009] A battery sorting system according to a second aspect of the present invention includes: (A) a first conveying unit that conveys waste including batteries; (B) a detection unit that detects batteries or waste products including batteries contained in the waste based on a transmission X-ray image of the waste being conveyed by the first conveying unit; (C) a second conveying unit that conveys waste from after the portion of the first conveying unit that is imaged with the transmission X-ray image to before the portion; (D) a transfer unit that, in response to an instruction from the detection unit, transfers a further portion of the waste, including batteries or waste products including batteries that have been conveyed by the first conveying unit and detected by the detection unit, to the second conveying unit; and (E) a discharge unit that discharges batteries or waste products including batteries being conveyed by the first conveying unit. The detection unit described above then determines whether the density, number or quantity of the detected battery or waste product containing a battery satisfies a first condition based on the transmitted X-ray image, and if it determines that the first condition is satisfied, it causes the detected battery or waste product containing a battery to be discharged to the discharge unit, and if it determines that the density, number or quantity of the detected battery or waste product containing a battery satisfies a second condition, it gives instructions to the transport unit. [Effects of the Invention]

[0010] According to one aspect, it becomes easier to separate batteries or waste products containing batteries from waste loads. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing the state of a battery mixed in waste. [Figure 2] FIG. 2 is a schematic top view of the battery sorting system according to the first embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of the configuration of the first control unit according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the configuration of the second control unit according to the first embodiment. [Figure 5A] FIG. 5A is a diagram for explaining the operation of the battery sorting system according to the first embodiment. [Figure 5B] FIG. 5B is a diagram for explaining the operation of the battery sorting system according to the first embodiment. [Figure 5C] FIG. 5C is a diagram for explaining the operation of the battery sorting system according to the first embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of the configuration of a first control unit in the second modification of the first embodiment. [Figure 7] FIG. 7 is a schematic top view of the battery sorting system according to the second embodiment. [Figure 8] FIG. 8 is a diagram illustrating a processing flow of the first control unit in the second embodiment. [Figure 9] FIG. 9 is a diagram illustrating a processing flow of the third control unit in the second embodiment. [Figure 10] FIG. 10 is a schematic top view of a battery sorting system according to the third embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of a functional configuration of a computer. DETAILED DESCRIPTION OF THE INVENTION

[0012] [Embodiment 1] In recycling facilities, large amounts of various waste materials (combustible, non-combustible, plastic, etc.) are often transported on a belt conveyor and fed into a crusher. For example, as shown schematically in Figure 1, waste material X is layered on top of the belt conveyor, and if battery B is mixed in, battery B cannot be seen from above the waste material X. Note that in Figure 1, side walls are provided on both sides of the belt conveyor, but for convenience of explanation, the side walls on the front side of the page are omitted, and only a partial cutout of the belt conveyor and waste material X is shown. In addition, the belt conveyor is assumed to transport waste material from right to left.

[0013] If waste X is irradiated with transmission X-rays on such a belt conveyor, the transmission X-rays will penetrate the waste X, and by analyzing the transmission X-ray image, it is possible to detect batteries that have been mixed in. However, only the position on the plane of the transmission X-ray image (here, XY coordinates) can be determined, and not the depth (Z coordinate). In such a situation, we would like to make it possible to properly sort batteries.

[0014] 2 shows a top view of the battery sorting system according to the first embodiment. The battery sorting system includes a belt conveyor 110 that transports waste X, including battery B; a first control unit 100 that captures and processes an X-ray image of the waste X on the belt conveyor 110; a push-out mechanism 120 that transfers the waste, including battery B, on the belt conveyor 110 to a belt conveyor 210 in response to instructions from the first control unit 100; a vibrator 130 that disperses the waste pushed out by the push-out mechanism 120; a belt conveyor 210 that transports the waste transferred from the belt conveyor 110; a second control unit 200 that captures and processes an X-ray image of the waste on the belt conveyor 210; and a sorting mechanism 220 that sorts and discharges batteries on the belt conveyor 210 in response to instructions from the second control unit 200.

[0015] The belt conveyor 110 is a type of conveying unit that conveys objects, and is configured to convey waste X from right to left, for example, as shown by arrow A. Waste that reaches the left end of the belt conveyor 110 can be crushed as is.

[0016] As shown in FIG. 3, the first control unit 100 includes a transmission X-ray imaging unit 101 that irradiates a predetermined range of the waste X transported by the belt conveyor 110 with transmission X-rays, for example from above, to capture a transmission X-ray image, a detection unit 102 that analyzes the transmission X-ray image to detect the battery B and identify its position (XY coordinates), etc., and an instruction output unit 103 that issues instructions to the extrusion mechanism 120 and the belt conveyor 110 based on the detection result of the detection unit 102.

[0017] The detection unit 102 is a trained model that has been machine-learned in advance using training data that identifies battery B in, for example, a transmission X-ray image. The instruction output unit 103 calculates the timing at which battery B and the surrounding waste will arrive in front of the push-out mechanism 120 based on the position of battery B that appears in the transmission X-ray image and the transport speed of the belt conveyor 110, and causes the push-out mechanism 120 to push out battery B and the surrounding waste toward the belt conveyor 210, and transfers them to the belt conveyor 210 via the vibrator 130. Note that the transport speed of the belt conveyor 110 may be slowed down or stopped at an appropriate timing so that the push-out mechanism 120 can reliably push out a portion of the waste including battery B.

[0018] The extrusion mechanism 120 is a type of transport unit that, in response to an instruction from the instruction output unit 103, horizontally cuts out a portion of the waste X containing the battery B as described above in a direction perpendicular to the traveling direction of the belt conveyor 110 and transfers it to the belt conveyor 210. Note that there is no problem if the waste X is mixed in after being cut out. Therefore, mechanisms such as a channelizer, flipper, drop-down, and up can also be used as an alternative to the extrusion mechanism 120.

[0019] The vibrator 130 is a type of mechanism for dispersing objects, and is provided to disperse the waste X and the battery B, and therefore may take the form of a vibrating sieve, a vibrating grizzly, a trommel, or the like, depending on the type and amount of the waste X. In some cases, the same effect can be achieved without providing the vibrator 130 by simply lowering the belt conveyor 210 below the belt conveyor 110 and providing a slope connecting the two.

[0020] The belt conveyor 210 is also a type of conveying unit that conveys objects, and is configured to convey waste X from right to left, for example, as shown by arrow C. Waste that reaches the left end of the belt conveyor 210 can be crushed as is.

[0021] As shown in FIG. 4, the second control unit 200 has a transmission X-ray imaging unit 201 that irradiates a predetermined range of a portion of the waste X transported by the belt conveyor 210 with transmission X-rays, for example from above, to capture a transmission X-ray image, a detection unit 202 that analyzes the transmission X-ray image to detect the battery B and identify its position (XY coordinates), etc., and an instruction output unit 203 that issues instructions to the sorting mechanism 220 and the belt conveyor 210 based on the detection results of the detection unit 202.

[0022] The detection unit 202 is a trained model that has been machine-learned in advance using training data that identifies battery B in, for example, a transmission X-ray image. The instruction output unit 203 outputs information that the sorting mechanism 220 uses to sort battery B.

[0023] For example, if the sorting mechanism 220 is a robotic hand, the detection unit 202 outputs the position (XY coordinates) of battery B within the range of motion of the robotic hand, the orientation, shape, and size of battery B identified from the transmitted X-ray image, and the like to the sorting mechanism 220, and the sorting mechanism 220 grasps and sorts battery B based on this information, for example, at a specified timing. If a waste product containing battery B is detected, the position, orientation, shape, and size of the waste product are used. If the type of waste product can be identified, that type may also be used. Furthermore, if multiple robotic hands are prepared and each is equipped with different attachments (for example, claws, suction devices, magnets, or a combination thereof), a robotic hand equipped with the optimal attachment may be selected based on the position, orientation, shape, size, etc. of battery B, etc., and the robotic hand may be made to grasp and sort battery B, etc.

[0024] When a robot hand or the like is used, if the robot hand or the like is compatible, the instruction output unit 203 may output a transmitted X-ray image or the like showing the battery B or the like detected by the detection unit 202, and the robot hand or the like may identify and select the battery B or the like on the belt conveyor 210 from the transmitted X-ray image.

[0025] Also, for example, if the sorting mechanism 220 is something like a paddle, the detection unit 202 outputs the position (XY coordinates) of battery B within the range of motion of the paddle to the sorting mechanism 220, and the sorting mechanism 220 flips and sorts battery B based on that information, for example, at a specified timing.

[0026] In some cases, workers may be used without providing the sorting mechanism 220. In such cases, the instruction output unit 203 may output the positions of the battery B and the like in the worker's work area to a monitor or indicate them on the belt conveyor 210 by a technique such as projection mapping, so that the worker can sort them manually.

[0027] In order to ensure that the sorting mechanism 220 can sort out a portion of the waste containing the battery B, the conveying speed of the belt conveyor 210 may be slowed down or stopped at an appropriate timing.

[0028] Next, an overview of the operation of the battery sorting system shown in Fig. 2 will be described with reference to Fig. 5A to Fig. 5C. As shown in Fig. 2, a battery B is mixed in a large amount of waste X, and is transported from the right end to the left on a belt conveyor 110.

[0029] When battery B is carried on the belt conveyor 110 to the imaging area of ​​the transmission X-ray imaging unit 101, the first control unit 100 detects battery B from the transmission X-ray image using the detection unit 102 and identifies its position, etc. Battery B is carried to the left on the belt conveyor 110 and, as shown in FIG. 5A, is carried in front of the push-out mechanism 120. At that timing, the instruction output unit 103 of the first control unit 100 instructs the push-out mechanism 120 to push out waste X on the belt conveyor 110 toward the belt conveyor 210.

[0030] In response to the instruction, the push-out mechanism 120 pushes a portion of the waste X, including the battery B, toward the belt conveyor 210 via the vibrator 130, as shown in FIG. 5B. The vibration of the vibrator 130 causes the pushed-out waste X to disperse and spread on the belt conveyor 210. This is expected to make the battery B (or a waste product containing it) visible. The dispersed waste X and battery B are transported from right to left by the belt conveyor 210.

[0031] When battery B is carried on the belt conveyor 210 to the imaging area of ​​the transmission X-ray imaging unit 201, the second control unit 200 detects battery B from the transmission X-ray image using the detection unit 202 and identifies its position, etc. Battery B is carried to the left on the belt conveyor 210 and, as shown in FIG. 5C , is carried to the operating range of the sorting mechanism 220. Furthermore, the instruction output unit 203 of the second control unit 200 instructs the sorting mechanism 220 to remove battery B from the belt conveyor 210. The instruction includes information such as the position, as described above.

[0032] In response to the instruction, the sorting mechanism 220 performs a sorting operation to identify the battery B, pick it up from the belt conveyor 210, and place it in a waste box or the like.

[0033] When waste X containing battery B is transported in front of the push-out mechanism 120, the transport speed of the belt conveyor 110 may be reduced or the belt conveyor 110 may be stopped. Similarly, when battery B is transported to the operating range of the sorting mechanism 220, the transport speed of the belt conveyor 210 may be reduced or the belt conveyor 210 may be stopped.

[0034] In this way, the battery B or the discarded product containing the battery B mixed in the large amount of waste X can be reliably sorted out.

[0035] [First Modification of First Embodiment] The detection unit 102 of the first control unit 100 and the detection unit 202 of the second control unit 200 are trained models that have been machine-learned in advance to detect battery B from, for example, a transmitted X-ray image, and in this case, typically also calculate a reliability score that the detected object is a battery. That is, an object that is highly likely to be estimated as a battery will have a high reliability score, and an object that is less likely to be estimated as a battery will have a low reliability score.

[0036] Therefore, the detection results can be changed by setting a detection threshold in advance and detecting only those objects with a reliability score above the detection threshold as batteries to be sorted out. Specifically, raising the detection threshold increases precision while decreasing recall, resulting in fewer false positives but more missed detections. Conversely, lowering the detection threshold decreases precision while increasing recall, resulting in more false positives but fewer missed detections.

[0037] Therefore, the detection thresholds in the detector 102 of the first control unit 100 and the detector 202 of the second control unit 200 may be set according to the purpose.

[0038] Specifically, if the waste disposal speed is given priority, the detection threshold is set high, and if the safety is given priority, the detection threshold is set low.

[0039] For the detection unit 102 of the first control unit 100, safety is prioritized and the detection threshold is set low to prevent missed detections, while for the detection unit 202 of the second control unit 200, since the battery is narrowed down to a certain extent by the detection unit 102 of the first control unit 100, the detection threshold may be set slightly higher with some consideration given to efficiency.

[0040] [Modification 2 of Embodiment 1] If Battery B is contained in a discarded product, it will be possible to change the operation accordingly if the type of discarded product can be identified. Also, if the risk of battery ignition can be estimated, it will be possible to change the operation accordingly.

[0041] In the present modified example 2, a first control unit 100a as shown in FIG.

[0042] The first control unit 100a includes a transmission X-ray imaging unit 1000, an image acquisition unit 1010, a pre-processing unit 1020, a first data storage unit 1030, an object detection unit 1040, a second data storage unit 1050, a detection control unit 1060, a plurality of first detection units 1070, a second detection unit 1080, an instruction output unit 1090, and a risk estimation unit 1100.

[0043] The transmission X-ray imaging section 1000 is similar to the transmission X-ray imaging section 101 .

[0044] The image acquisition unit 1010 periodically acquires transmission X-ray images from the transmission X-ray imaging unit 1000, for example, and outputs them to the preprocessing unit 1020. The preprocessing unit 1020 performs preprocessing such as resizing the transmission X-ray images and inverting the grayscale for the following processing, and stores the images in the first data storage unit 1030. Note that the inversion of the grayscale is performed to improve the accuracy of battery detection in later processing.

[0045] The object detection unit 1040 executes a process for detecting image portions in which any of a plurality of predetermined types of waste products appear in the preprocessed transmission X-ray image stored in the first data storage unit 1030. The object detection unit 1040 is, for example, a single shot multibox detector (SSD). When used, a single shot detector locates the positions of multiple objects in an image, encloses them in rectangles (i.e., bounding boxes), and further identifies the objects for each rectangle. For details, see Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, CY, & Berg, AC (2016, October). SSD: Single shot multibox detector. In European conference on computer vision (pp. 21-37). Springer, Cham., etc. In this embodiment, an SSD is used that has been trained using training data in which a rectangle surrounding each waste product sample and its product name are added in a transmission X-ray image of a predetermined number of waste product samples. It may also be possible to detect a predetermined number of waste products by adjusting the weights of an existing SSD or by performing transfer learning. The processing results of the object detection unit 1040 are stored in the second data storage unit 1050.

[0046] The detection control unit 1060 causes the plurality of first detection units 1070 and second detection units 1080 to execute processing for battery detection using the data stored in the first data storage unit 1030 and the second data storage unit 1050. Specifically, for image portions in which the type of discarded product has been identified by the object detection unit 1040, the detection control unit 1060 causes the first detection unit 1070 corresponding to the type of discarded product to execute processing for battery detection, and for image portions in which the type of discarded product has not been identified, causes the second detection unit 1080 to execute processing for battery detection.

[0047] Furthermore, the detection control unit 1060 stores data currently being processed in the second data storage unit 1050, and outputs the battery detection result to the instruction output unit 1090. The instruction output unit 1090 outputs an instruction to the extrusion mechanism 120 based on the battery detection result. This is similar to the instruction output unit 103.

[0048] Each of the multiple first detectors 1070 specializes in a different type of discarded product and detects the batteries built into the discarded product in its specialized type in a transmission X-ray image of the discarded product. Each first detector 1070 is, for example, an SSD, trained using training data in which a rectangle enclosing the built-in battery is added in a transmission X-ray image of a discarded product sample of the type in which the first detector 1070 specializes. It should be noted that an existing SSD may be made to detect built-in batteries by adjusting the weights or performing transfer learning. The processing results of each first detector 1070 are output to the detection control unit 1060.

[0049] Furthermore, unlike the first detection unit 1070, the second detection unit 1080 is a general-purpose detection unit that detects batteries that appear in transmission X-ray images, rather than batteries built into specific types of waste products. The second detection unit 1080 is also, for example, an SSD, and is an SSD that has been trained using training data in which a rectangle surrounding a battery is added in a transmission X-ray image showing waste product samples, a single battery, or waste in which a single battery is mixed with metals, etc. Note that it may be possible to detect batteries by adjusting the weights or performing transfer learning on an existing SSD. The processing results of the second detection unit 1080 are output to the detection control unit 1060.

[0050] Furthermore, when the first detection unit 1070 and the second detection unit 1080 detect a battery, the risk estimation unit 1100 executes a process of estimating the risk of fire of the battery.

[0051] The risk of fire is based on, for example, a risk level α for battery capacity, a risk level β for battery type, a risk level γ for type of discarded product, and a risk level η for battery size.

[0052] The risk level α for battery capacity is calculated by, for example, estimating the battery capacity of a battery appearing in a radiographic X-ray image using a trained model that has been machine-learned in advance using training data that associates a radiographic X-ray image of the battery with the battery capacity, and deriving the risk level α from the battery capacity. For example, if the battery capacity is estimated to be large, the risk level α is estimated to be high. Note that instead of a trained model, some kind of rule-based algorithm may be adopted.

[0053] The risk level β for the battery type is calculated by, for example, estimating the battery type for a battery that appears in a radiographic X-ray image using a trained model that has been machine-learned in advance using training data that associates a radiographic X-ray image of the battery with the battery type, and deriving a predetermined risk level β according to the battery type. For example, if the battery is a lithium-ion battery, the risk level β is estimated to be high. Note that instead of a trained model, some kind of rule-based algorithm may be used.

[0054] Regarding the risk level γ for the type of waste product, for example, when the type of waste product is identified by the object detection unit 1040, a risk level γ that is preset according to the type of waste product is output. For example, it is estimated that the risk level of a mobile battery is high.

[0055] The risk η for battery size is calculated by, for example, multiplying the area of ​​the battery in the transmitted X-ray image by the thickness estimated from the brightness value using a preset conversion formula to calculate the volume, and then deriving the risk η from the volume based on, for example, a predetermined formula. For example, it is estimated that the risk η is higher as the volume of the battery increases.

[0056] Taking all of this into consideration, the fire risk can be expressed as follows: Ignition risk = α + β + γ + η However, instead of calculating all the risk levels, it is also possible to calculate only some of the risk levels, in which case the risk levels that have not been calculated can be set to 0.

[0057] Furthermore, when the reliability scores are obtained from the first detection unit 1070 and the second detection unit 1080, the fire risk may be calculated as reliability×(α+β+γ+η).

[0058] When such a risk of fire is obtained, for example, the battery or waste product containing the battery whose risk of fire is above a threshold value is pushed out toward the belt conveyor 210 by the pushing mechanism 120 as the battery or waste product containing the battery to be discharged.

[0059] Furthermore, if there is a battery or a waste product containing a battery whose ignition risk level is equal to or higher than the second threshold, the conveying speed of the belt conveyor 110 may be reduced, or if the battery or a waste product containing a battery has moved in front of the push-out mechanism 120, the belt conveyor 110 may be stopped and then pushed out by the push-out mechanism 120. This makes it possible to reliably push out the battery or the waste product containing a battery toward the belt conveyor 210.

[0060] It should be noted that not only the first control unit 100 but also the second control unit 200 may employ the configuration shown in FIG. 6 to calculate the risk of fire.

[0061] When such a risk of fire is obtained, for example, a battery or a waste product containing a battery whose risk of fire is equal to or greater than a threshold value is sorted by the sorting mechanism 220 as a battery to be sorted or a waste product containing the battery.

[0062] Furthermore, if there is a battery or a waste product containing a battery whose ignition risk is equal to or higher than the second threshold, the transport speed of the belt conveyor 210 may be reduced, or if the battery or a waste product containing a battery has moved into the operating range of the sorting mechanism 220, the belt conveyor 210 may be stopped before the sorting mechanism 220 sorts the battery or the waste product containing a battery. This allows the battery or the waste product containing a battery to be sorted reliably.

[0063] Furthermore, if multiple types of sorting methods are available, the sorting method may be selected according to the fire risk. For example, if the fire risk is higher than a threshold, sorting may be performed manually by an operator, and if the fire risk is lower than the threshold, sorting may be performed by the automatic sorting mechanism 220. Furthermore, a threshold may be set so that the automatic sorting mechanism 220 is used selectively.

[0064] [Third Modification of First Embodiment] In the second modification, if a battery is included in a discarded product and the discarded product is one of a plurality of predetermined types, the type is identified. The type may be used to change the operation.

[0065] For example, the push-out mechanism 120 may be configured to push out pre-specified types of waste products with high resource value toward the belt conveyor 210.

[0066] Similarly, the sorting mechanism 220 may be configured to sort out waste products of a type that has a high predetermined resource value.

[0067] Furthermore, the first control unit 100 and the second control unit 200 may use different indices to detect batteries that should be discharged.

[0068] [Fourth Modification of First Embodiment] The reliability score, the fire risk, and the type of waste product may not be handled separately, but may be arbitrarily combined to change the operation.

[0069] For example, if any of the indicators satisfies the condition, the corresponding action may be performed, or if all of the indicators satisfy the corresponding conditions, the corresponding action may be performed.

[0070] [Fifth Modification of First Embodiment] In the first embodiment, the second control unit 200 has an example of including the transmitted X-ray imaging unit 201, similar to the first control unit 100. However, if the waste is properly dispersed on the belt conveyor 210, a visible light camera or an infrared camera may also be used. In this case, the detection unit 202 is capable of determining whether or not a battery is present from the visible light image or infrared image output by the camera. In this case, it is not possible to determine whether or not a battery is present in a waste product, but it is possible to treat certain types of waste products as if they contain a battery.

[0071] [Embodiment 2] A top view of a battery sorting system according to the second embodiment will be outlined with reference to FIG.

[0072] The battery sorting system of the second embodiment includes a belt conveyor 110, a first control unit 100, a push-out mechanism 120, a vibrator 130, a belt conveyor 210, a third control unit 300, a push-out mechanism 330, a push-out mechanism 340, a disposal box 320, and a belt conveyor 310.

[0073] The belt conveyor 110, the first control unit 100, the push-out mechanism 120, and the vibrator 130 are the same as those in the first embodiment. That is, the batteries or waste products containing batteries detected by the first control unit 100 are pushed by the push-out mechanism 120 toward the belt conveyor 210 via the vibrator 130, and are transferred onto the belt conveyor 210 in a dispersed manner.

[0074] The belt conveyor 210 transports waste and the like from right to left as shown by arrow C. The third control unit 300 takes a transmitted X-ray image of the waste and the like on the belt conveyor 210 and executes a process of detecting the battery.

[0075] In this embodiment, for example, the number of batteries detected in the imaging area is counted, and if the number of batteries is zero, the batteries may be discarded as is, and the batteries are simply passed down the belt conveyor 210 without operating the push-out mechanisms 330 and 340. On the other hand, if the number of detected batteries is one or more but less than a predetermined number, the push-out mechanism 340 is operated to push out and transfer the waste or the like corresponding to the imaging area toward the belt conveyor 310. The belt conveyor 310 is a U-shaped belt conveyor that transports waste from the rear side to the front side of the third control unit 300, and transports it again upstream of the belt conveyor 210 via the vibrator 350.

[0076] By repeating this process, an area with gradually increasing battery density is formed on the belt conveyor 210. Therefore, when the number of detected batteries is equal to or greater than a predetermined number, the push-out mechanism 340 is not operated, and the push-out mechanism 330 is operated to push out waste or the like corresponding to the image capture area into the waste box 320.

[0077] The conveying speed of the belt conveyor 210 may be reduced or the belt conveyor 210 may be stopped when the push-out mechanisms 330 and 340 are operating.

[0078] Also, although an example of evaluating the number of batteries detected in the imaging area has been shown, it is also possible to evaluate the density (e.g., projected area of ​​the battery / area of ​​the imaging area), quantity (e.g., projected area of ​​the battery), etc.

[0079] 8 and 9, the operation of the battery sorting system according to this embodiment will be summarized. The first control unit 100 takes a transmission X-ray image of waste or the like on the belt conveyor 110 (FIG. 8: step S1). Then, the first control unit 100 executes a battery detection process on the taken transmission X-ray image (step S3). Here, if a battery (or a waste product containing a battery) cannot be detected in the transmission X-ray image (step S5: No route), the process proceeds to step S11. That is, the waste is discarded as is.

[0080] On the other hand, if a battery is detected, the first control unit 100 instructs the push-out mechanism 120 to push the waste including the battery onto the next belt conveyor 210 in accordance with the movement on the belt conveyor 110 (step S7). In response to this instruction, the push-out mechanism 120 pushes out the waste, etc., and transfers the waste, etc., while dispersing it onto the next belt conveyor 210 (step S9).

[0081] Then, steps S1 to S9 are repeated until the process is completed.

[0082] In response to this, the third control unit 300 takes a transmission X-ray image of the waste, etc. on the belt conveyor 210 (FIG. 9: step S21). Then, the third control unit 300 executes a battery detection process on the taken transmission X-ray image (step S23). Here, if a battery (or a waste product including a battery) cannot be detected in the transmission X-ray image (step S25: No route), the process proceeds to step S37. As a result, this portion of waste is disposed of as is.

[0083] On the other hand, if batteries are detected, the third control unit 300 determines whether the number of detected batteries is less than a predetermined value (step S27). If this condition is met, the third control unit 300 instructs the push-out mechanism 340 to push the waste containing batteries onto the return belt conveyor 310 in accordance with the movement on the belt conveyor 210 (step S29). In response to the instruction, the push-out mechanism 340 pushes out the waste, etc., and transfers it to the return belt conveyor 310 (step S31). Then, the process proceeds to step S37. This returns the waste containing batteries to the front of the belt conveyor 210.

[0084] On the other hand, if the detected number of batteries is equal to or greater than the predetermined value, the third control unit 300 instructs the push-out mechanism 330 to push the waste including the batteries into the waste box 320 in accordance with the movement on the belt conveyor 210 (step S33). In response to the instruction, the push-out mechanism 330 pushes and discharges the waste including the batteries into the waste box 320 (step S35). Then, the process proceeds to step S37. This allows batteries that are relatively closely packed to be placed into the waste box 320 all at once.

[0085] This process, that is, steps S21 to S37, is repeated until the process is completed.

[0086] By adopting the battery sorting system described above, it will be possible to sort batteries contained in large amounts of waste.

[0087] [Modification of the second embodiment] The first control unit 100 is the same as that in the first embodiment, and therefore the matters explained in the modified example of the first embodiment can be applied as they are.

[0088] The third control unit 300 may also determine whether a battery or waste product should be sorted out based on the reliability, the risk of fire, and the type of waste product including the battery.

[0089] [Embodiment 3] In the first and second embodiments, it is assumed that a large amount of waste is placed on the belt conveyor 110 for processing. That is, the belt conveyor 110, the first control unit 100, the push-out mechanism 120, and the vibrator 130 are used to perform a preliminary processing step of extracting a portion of the waste, including batteries, to be discharged, from the large amount of waste.

[0090] On the other hand, if it is not necessary to process a large amount of waste, the waste may be placed on the belt conveyor 210 from the beginning, without using the belt conveyor 110, the first control unit 100, the push-out mechanism 120, and the vibrator 130. In other words, the system may be operated as shown in FIG.

[0091] 2, the belt conveyor 110, the first control unit 100, the pushing mechanism 120, and the vibrator 130 may be omitted.

[0092] Although the embodiments of the present invention have been described above, the present invention is not limited thereto. For example, the functional configuration examples of the first control units 100 and 100a, the second control unit 200, etc. are merely examples. The processing flows shown in FIGS. 8 and 9 are also merely examples, and the processing order may be changed or multiple steps may be executed simultaneously as long as similar processing results can be obtained. Furthermore, while the above description uses a trained model using machine learning as an example, appropriate technology related to artificial intelligence may also be employed for implementation. Furthermore, in some cases, the first embodiment may be configured such that batteries and other waste extruded by the extrusion mechanism 120 are manually sorted by workers without using the second control unit 200 and the sorting mechanism 220.

[0093] Furthermore, the components other than the transmission X-ray imaging unit of the first control unit 100 or 100a, second control unit 200, or third control unit 300 described above are, for example, a computer device as shown in FIG. 11 , in which a memory 2501, a CPU 2503, a hard disk drive (HDD) 2505, a display control unit 2507 connected to a display device 2509, a drive device 2513 for a removable disk 2511, an input device 2515, and a communication control unit 2517 for connecting to a network are connected via a bus 2519. An operating system (OS) and application programs for carrying out the processes of this embodiment are stored in the HDD 2505, and are read from the HDD 2505 to the memory 2501 when executed by the CPU 2503. The CPU 2503 controls the display control unit 2507, the communication control unit 2517, and the drive device 2513 according to the processing content of the application program to perform predetermined operations. Furthermore, data during processing is primarily stored in memory 2501, but may also be stored in HDD 2505. In an embodiment of the present invention, an application program for performing the above-described processing is distributed by being stored on a computer-readable removable disk 2511, and is installed on HDD 2505 from drive device 2513. It may also be installed on HDD 2505 via a network such as the Internet and communication control unit 2517. Such a computer device realizes the various functions described above through organic cooperation between hardware such as the CPU 2503 and memory 2501 described above and programs such as the OS and application programs.

[0094] Furthermore, the detection units 102 and 202, the object detection unit 1040, the first battery detection unit 1070, and the second battery detection unit 1080 may be implemented using multiple GPUs (Graphics Processing Units) instead of the CPU 2503. Furthermore, the HDD 2505 may be another storage device, such as a solid state drive.

[0095] Furthermore, instead of implementing all of the functions of the first control unit 100 or 100a, the second control unit 200, or the third control unit 300 (excluding the transmitted X-ray image capturing unit) in a single computer device, the functions may be shared among multiple computer devices. For example, multiple functions may be implemented in different computers connected to a network. Also, these may be implemented on the cloud, and in the recycling plant, only the parts that utilize it may be implemented in the control unit. Note that whether they are implemented in a single computer device or in multiple computer devices, the entire system is referred to as an information processing system.

[0096] The above-described embodiment can be summarized as follows.

[0097] The battery sorting system according to the first aspect of this embodiment includes: (A) first and second conveying units (e.g., belt conveyors 110 and 120) that transport waste containing batteries; (B) a first detection unit (e.g., first control units 100 and 100a) that detects batteries or waste products containing batteries contained in the waste based on a first transmitted X-ray image of the waste being transported by the first conveying unit; (C) a transfer unit (e.g., extrusion mechanism 120) that, when a battery or a waste product containing a battery is detected by the first detection unit, transfers a portion of the waste containing the battery or the waste product containing the battery to the second conveying unit; and (D) a second detection unit (e.g., second control unit 200, third control unit 300) that detects batteries or waste products containing batteries contained in the portion of the waste based on a second image of the portion of the waste being transported by the second conveying unit, and outputs information for sorting the detected batteries or waste products containing batteries.

[0098] This two-stage sorting process makes it easier to separate batteries even when they are mixed in with a large amount of waste. The information used for sorting can be output to the device or to the worker.

[0099] The battery sorting system described above may further include (E) a sorting unit that sorts batteries or waste products containing batteries that are transported by the second transport unit and detected by the second detection unit based on the information from the second detection unit. This shows a configuration for automatic sorting.

[0100] The first detection unit may estimate the type of discarded product containing a battery, and if the type is a predetermined type, detect the discarded product as containing a battery, since discarded products may also have value.

[0101] Furthermore, the first detection unit described above may estimate the risk of fire of the battery or a discarded product containing a battery, and may detect the battery or a discarded product containing a battery if the risk of fire is equal to or greater than a predetermined value, taking safety into consideration.

[0102] Furthermore, the second detection unit may estimate the type of discarded product containing a battery, and if the type is a predetermined type, detect the discarded product as containing a battery, since discarded products may also have value.

[0103] Furthermore, the second detection unit described above may estimate the risk of fire of the battery or a discarded product containing a battery, and may detect the battery or a discarded product containing a battery if the risk of fire is equal to or greater than a predetermined value, taking safety into consideration.

[0104] Furthermore, the second detection unit may estimate the fire risk of the battery or the waste product containing the battery and identify a sorting method to be adopted depending on the fire risk. In this case, the information may include information about the identified sorting method. For example, the sorting method may include, depending on the fire risk, which type of sorting unit to select, what the conveying speed of the conveying unit should be, whether to have a worker sort the batteries, etc.

[0105] In the battery sorting system described above, it is preferable to disperse the portion of the waste when transferring the portion of the waste to the second transport unit. For this purpose, a vibrating unit or a slope may be provided.

[0106] The battery sorting system described above may further include (F) a third conveying unit (e.g., belt conveyor 310) that conveys waste from after the second image capturing portion in the second conveying unit to before the second image capturing portion, (G) a second transfer unit (e.g., extrusion mechanism 340) that transfers a further portion of the waste, including batteries or waste products including batteries that have been conveyed by the second conveying unit and detected by the second detection unit, to the third conveying unit in response to instructions from the second detection unit, and (H) a discharge unit (e.g., extrusion mechanism 330) that discharges batteries or waste products including batteries that are being conveyed by the second conveying unit. In this case, the second detection unit (for example, the third control unit 300) described above may determine, based on the second image, whether the density, number, or amount of the detected batteries or waste products containing batteries satisfies the first condition, and if it determines that the first condition is satisfied, output the information to the discharge unit to have the detected batteries or waste products containing batteries discharged to the discharge unit, and if it determines that the density, number, or amount of the detected batteries or waste products containing batteries satisfies the second condition, give an instruction to the second transport unit. In this way, the batteries will gather in a relatively dense form in the discharge unit, making them easier to discharge.

[0107] The battery sorting system according to the second aspect of this embodiment includes: (A) a first conveying unit (e.g., belt conveyor 210) that conveys waste containing batteries; (B) a detection unit (e.g., third control unit 300) that detects batteries or waste products containing batteries contained in the waste based on a transmission X-ray image of the waste being conveyed by the first conveying unit; (C) a second conveying unit (e.g., belt conveyor 310) that conveys waste from after the portion of the first conveying unit where the transmission X-ray image is captured to before the portion; (D) a transfer unit (e.g., extrusion mechanism 340) that transfers a further portion of the waste, including batteries or waste products containing batteries that have been conveyed by the first conveying unit and detected by the detection unit, to the second conveying unit in response to instructions from the detection unit; and (E) a discharge unit (e.g., extrusion mechanism 330) that discharges batteries or waste products containing batteries being conveyed by the first conveying unit. The detection unit described above then determines whether the density, number or quantity of the detected battery or waste product containing a battery satisfies a first condition based on the transmitted X-ray image, and if it determines that the first condition is satisfied, it causes the detected battery or waste product containing a battery to be discharged to the discharge unit, and if it determines that the density, number or quantity of the detected battery or waste product containing a battery satisfies a second condition, it gives instructions to the transport unit.

[0108] If the amount of waste to be processed is not too large, this configuration also makes it easier to sort the batteries.

[0109] A program for causing a processor to perform the above-described processing can be created, and the program is stored in a computer-readable storage medium or storage device, such as a flexible disk, an optical disk such as a CD-ROM (Read Only Memory), a magneto-optical disk, a semiconductor memory (e.g., a ROM), a hard disk, etc. Data during processing is temporarily stored in a storage device such as RAM (Random Access Memory). [Explanation of symbols]

[0110] 100, 100a First control section 200 Second Control Section 300 Third Control Section 110,210,310 conveyor belt 120,330,340 Extrusion mechanism 220 Sorting Mechanism

Claims

1. first and second conveying units for conveying waste including batteries; a first detection unit that detects batteries or waste products containing batteries contained in the waste based on a first transmission X-ray image of the waste being transported by the first transport unit; a transfer unit that transfers a portion of waste including the battery or the waste product including the battery to the second transport unit when the first detection unit detects the battery or the waste product including the battery; a second detection unit that detects batteries or waste products containing batteries contained in the portion of waste based on a second image of the portion of waste being transported by the second transport unit, and outputs information for sorting the detected batteries or waste products containing batteries; A battery sorting system having:

2. a sorting unit that sorts the battery or a waste product containing a battery that is transported by the second transport unit and detected by the second detection unit based on the information from the second detection unit.

10. The battery sorting system of claim 1, further comprising:

3. The first detection unit A type of a discarded product including the battery is estimated, and if the type is a predetermined type, the discarded product including the battery is detected. The battery sorting system of claim 1 .

4. The first detection unit A fire risk of the battery or a discarded product including a battery is estimated, and when the fire risk is equal to or greater than a predetermined value, the battery or the discarded product including a battery is detected. The battery sorting system of claim 1 .

5. The second detection unit A type of a discarded product including the battery is estimated, and if the type is a predetermined type, the discarded product including the battery is detected. The battery sorting system of claim 1 .

6. The second detection unit A fire risk of the battery or a discarded product including a battery is estimated, and when the fire risk is equal to or greater than a predetermined value, the battery or the discarded product including a battery is detected. The battery sorting system of claim 1 .

7. The second detection unit Estimating the fire risk of the battery or the waste product containing the battery, and specifying a sorting method to be adopted depending on the fire risk; The information includes information about the specified selection manner. The battery sorting system of claim 1 .

8. Dispersing the portion of the waste material when transferring the portion of the waste material to the second transport section. The battery sorting system of claim 1 .

9. a third transport unit that transports waste from behind the capture portion of the second image in front of the capture portion of the second image in the second transport unit; a second transfer unit that transfers a further portion of the waste, including the battery or the waste product including the battery, which is transported by the second transport unit and detected by the second detection unit, to the third transport unit in response to an instruction from the second detection unit; a discharge section that discharges the batteries or waste products including batteries being transported in the second transport section; and The second detection unit determining whether the density, number, or amount of the detected batteries or waste products containing batteries satisfies a first condition based on the second image; When it is determined that the first condition is satisfied, the information is output to the discharge unit, and the detected battery or a waste product including the battery is discharged to the discharge unit; If it is determined that the density, number, or amount of the detected batteries or waste products containing batteries satisfies the second condition, an instruction is given to the second transport unit. The battery sorting system of claim 1 .

10. a first transport unit that transports waste including batteries; a detection unit that detects batteries or waste products containing batteries contained in the waste based on a transmission X-ray image of the waste being transported by the first transport unit; a second transport unit that transports waste from a rear portion of the first transport unit to a front portion of the first transport unit where the transmission X-ray image is captured; a transfer unit that transfers a further portion of the waste, including the battery or the waste product including the battery, which is transported by the first transport unit and detected by the detection unit, to the second transport unit in response to an instruction from the detection unit; a discharge section that discharges the batteries or waste products including batteries being transported in the first transport section; and The detection unit determining whether the density, number, or amount of the detected batteries or waste products containing batteries satisfies a first condition based on the transmitted X-ray image; When it is determined that the first condition is satisfied, the detected battery or a waste product including the battery is discharged to the discharge unit; If it is determined that the density, number, or amount of the detected batteries or waste products containing batteries satisfies the second condition, an instruction is given to the transport unit. Battery sorting system.

Images