Scrap conveying device, scrap conveying method, program, and recording medium

Abstract

Provided is a technology capable of sorting and taking out scrap and transporting the scrap to another location. This scrap transport device selects and transports a scrap group (1) composed of a plurality of pieces of scrap (2), the scrap transport device comprising: an imaging device (3) that images the scrap group (1); an image analysis unit (4) that, when considering a collection of one or more pieces of scrap (2) present in a region among one or more regions located in the scrap group (1) as detected scrap, analyzes the captured image captured by the imaging device (3) and estimates scrap information that includes sorting information that specifies the location of the detected scrap and the content of the pieces of scrap (2) that constitute the detected scrap; a selection calculation unit (5) that refers to the estimated selection information and selects the detected scrap that satisfies a preset selection condition; and a transfer device (6) that takes out and transports the pieces of scrap (2) from the region in which the detected scrap selected by the selection calculation unit (5) is located.

Description

[Technical Field]

[0001] The present invention relates to a technique for transporting scrap selected from a plurality of scraps. [Background technology]

[0002] Conventionally, there have been techniques for partially removing scrap from a plurality of scraps and transporting the scraps, such as those described in Patent Documents 1 and 2. In this specification, "taking out" is also referred to as "picking up." In Patent Document 1, iron-based scrap is picked up and transported by a lifting magnet. At this time, Patent Document 1 also describes that the lifted scrap is analyzed for its components by a component analyzer contained within the lifting magnet. Patent Document 2 also describes a method of lifting and transporting some of the piled up scrap iron using a lifting tool. Patent Document 2 also describes a method of photographing a collection of scrap iron containing a mixture of multiple grades of scrap iron according to size. The photographed image is then analyzed to estimate the proportion of each grade contained in the photographed collection of scrap iron. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2022-166489 [Patent Document 2] Japanese Patent Application Publication No. 2020-95709 Summary of the Invention [Problem to be solved by the invention]

[0004] For example, in a steelworks, an electric furnace can be an example of a destination for iron-based scrap picked from a plurality of iron-based scraps. Here, the target composition of the scrap to be fed into the electric furnace is predetermined. Therefore, it is necessary to adjust the composition and amount of the scrap so that it meets the target composition and then discharge the scrap to the electric furnace.

[0005] However, in Patent Document 1, the components of the picked-up scrap are analyzed. However, this analysis is performed after the scrap is picked up. This makes it difficult to selectively remove and transport scrap from multiple scraps based on their components. In other words, it is difficult to selectively discharge scrap based on its components. Therefore, if the technology in Patent Document 1 is applied to discharging scrap to an electric furnace, there is a risk that the accuracy of the scrap discharged to the electric furnace may be reduced accordingly. In addition, in Patent Document 2, the grade ratio of the entire collection of piled or picked-up iron scrap is estimated from an image of the collection. However, it is difficult to select and transport only scrap that meets certain conditions from the collection of iron scrap in the image. Therefore, when the technology in Patent Document 2 is applied to discharging scrap into an electric furnace, there is a problem in that it is difficult to selectively discharge scrap according to its composition.

[0006] The present invention has been made in light of the above-mentioned points, and aims to provide a technology that allows the target scrap to be selected, removed, and transported to another location. [Means for solving the problem]

[0007] In order to solve the problem, one aspect of the present invention is a scrap conveying device that selects and conveys scrap from a scrap group consisting of multiple scraps, and includes an imaging device that images the scrap group, an image analysis unit that analyzes the image captured by the imaging device and estimates scrap information for each detected scrap, including sorting information that identifies the contents of the scrap that make up the detected scrap, when a collection of one or more scraps present in each of one or more areas located in the scrap group is defined as detected scrap, a selection calculation unit that refers to the estimated sorting information and selects detected scrap that satisfies predetermined selection conditions, and a transfer device that removes and conveys scrap from the area where the detected scrap selected by the selection calculation unit is located.

[0008] Here, the detected scrap may be made up of one scrap, or may be made up of a set of multiple scraps. [Effects of the Invention]

[0009] According to this aspect of the present invention, before the scrap to be transported is removed, the sorting information of each detected scrap that constitutes the scrap group is acquired, and then the scrap to be transported is selected based on the sorting information, and the selected scrap can be transported. Therefore, according to this aspect of the present invention, it is possible to pick up and transport only scrap that mainly satisfies the selection conditions without analyzing the components of the scrap after picking it up. Here, the present invention also includes a case where a plurality of detected scraps are combined to satisfy the selection conditions. According to an aspect of the present invention, for example, when assuming discharge to an electric furnace, the selection conditions can be set to a target composition, and scrap having the target composition or a composition close to the target composition can be selectively transported. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram showing an example of the configuration of a scrap transport device according to an embodiment of the present invention; [Figure 2] FIG. 10 is a diagram illustrating an example of setting an area. [Figure 3] FIG. 10 is a diagram illustrating an example of a process flow of a scrap transport method. [Figure 4] 1A and 1B are diagrams illustrating an example of scrap transportation. FIG. 1A is a diagram illustrating scraps A and B selected for transportation. FIG. 1B is a diagram illustrating scrap A being transported by a transport actuator. FIG. 1C is a diagram illustrating scraps A and B being transported together. [Figure 5] 10A and 10B are diagrams illustrating an example of repeated scrap transport. [Figure 6] 10A and 10B are diagrams illustrating an example of continuous lifting of scrap. [Figure 7] 10A and 10B are diagrams illustrating an example of continuous lifting of scrap. [Figure 8] FIG. 10 is a diagram illustrating an example of adjusting the lifting amount of scrap. [Figure 9] FIG. 10 is a diagram illustrating an example of a program and a medium storing the program. DETAILED DESCRIPTION OF THE INVENTION

[0011] Next, an embodiment of the present invention will be described with reference to the drawings. The scrap transport device and transport method of this embodiment transport a plurality of scraps 2 placed on a loading location such as the floor of a yard or the bed of a truck. This embodiment is a transport device and transport method that picks up scraps 2 selected from a scrap group 1 made up of the plurality of scraps 2. The picked-up scraps 2 are transported to a location different from the above-mentioned loading location. The scrap pile 1 is made up of a collection of multiple scraps 2 placed on a placement location. The scrap pile 1 is usually made up of scraps 2 piled up one on top of another. The scrap pile 1 is placed, for example, in a heap (see FIG. 1). The scrap pile 1 is also placed, for example, with its top surface flat (see FIG. 7).

[0012] The scrap 2 to be transported in the present invention is not particularly limited in terms of material or shape. The present invention is a technology suitable for metal scrap. For example, the present invention can be particularly suitable for selecting scrap 2 to be fed into a furnace (such as an electric furnace or a converter) in a steelworks. In this embodiment, the scrap 2 will be described as an example of iron-based scrap. As shown in FIG. 1, the scrap transport device of this embodiment includes an imaging device 3, an image analysis unit 4, a selection calculation unit 5, and a transfer device 6.

[0013] <Imaging device 3> The imaging device 3 is a device that captures an image of the scrap pile 1 placed in a placement location. The target scrap pile 1 is usually in a piled state (see FIG. 1). Note that the image captured by the imaging device 3 may also capture something other than the scraps 2, such as background objects around the scrap pile 1. The imaging device 3 may also be configured to capture an image of a collection of scraps 2 that are part of the scrap pile 1. In this example, the collection of scraps 2 captured by the imaging device 3 is also called the scrap group 1. The imaging device 3 is, for example, a camera. The camera captures an image of the scrap group 1 from above. The imaging device 3 may be configured with one camera or multiple cameras.

[0014] <Image Analysis Unit 4> Here, in this specification, a detected scrap is defined as a collection of one or more scraps 2 present in one or more regions of a scrap group 1. The regions may be set continuously or discretely. Furthermore, the regions may be arranged either fixedly or dynamically relative to the image. When one scrap 2 constitutes one detected scrap, for example, the position (outline) of each scrap 2 in the image is dynamically set as the above-mentioned region. When multiple scraps 2 constitute one detected scrap, for example, the region is set first. For example, as shown in FIG. 2, an image region 10 is divided into multiple regions 11, and the scraps 2 present in each divided region 11 are defined as detected scraps for the target region 11. As shown in FIG. 1, the image analysis unit 4 includes an image acquisition unit 4A and a scrap information estimation unit 4B.

[0015] [Image acquisition unit 4A] The image acquisition unit 4A acquires the captured image of the scrap group 1 captured by the imaging device 3. The image acquisition unit 4A, for example, sends an imaging command to the imaging device 3 and acquires the captured image captured by the imaging device 3 in response to the imaging command.

[0016] [Scrap Information Estimation Section 4B] The scrap information estimation unit 4B analyzes the captured images captured by the imaging device 3 and performs a process to estimate scrap information including selection information that identifies the position of each detected scrap and the contents of the scraps 2 that make up the detected scrap. The sorting information is information for identifying scrap 2 to be picked up (taken out) from the scrap group 1. The sorting information is determined and set according to the pick-up conditions.

[0017] In the following description, a case where scrap 2 to be fed into a furnace such as an electric furnace is selected will be described as an example. In this case, the composition of the scrap 2 to be input is important. For this reason, the sorting information will be explained using composition estimation information that can estimate the composition of the detected scrap as an example. In this case, the selection conditions described below will be target composition information. The component estimation information may be the component itself of the scrap 2, or may be information that enables the component to be estimated. Examples of information that enables the component to be estimated include the type of scrap 2 and the grade of the scrap 2. The type is information indicating the origin of the scrap 2. The type is the variety of the scrap 2, such as wire material, automobile panel material, etc. The type is classification information that can identify the material. The grade is classification information that is set according to the material and size of the scrap 2.

[0018] The image analysis unit 4 of this embodiment further includes a weight estimation unit 4 C. The weight estimation unit 4 C may be omitted. The weight estimation unit 4C obtains estimated weight information of the detected scrap as the scrap information. The scrap information may also include estimated information on the solubility of the detected scrap. The solubility is an index of the degree to which scrap 2 melts when heated under the same conditions. When scrap 2 is discharged to an electric furnace or the like, the solubility of scrap 2 in the furnace is also an important factor for the operation of the furnace. The solubility is affected by the size and shape of scrap 2. This solubility may be used as one of the sorting information.

[0019] A method for estimating the sorting information of each detected scrap from a captured image involves, for example, performing machine learning using a large amount of training data to generate a trained model in advance. The training data includes, for example, images of one or more scraps 2 and sorting information for the scraps 2 in the images. The generated trained model uses, for example, the captured image as input data and the sorting information of the detected scraps in each region of the captured image as output data. Note that, as described above, the regions may be statically or dynamically arranged. The scrap information estimation unit 4B then inputs the captured image into the trained model to obtain the sorting information for each detected scrap. Since the scrap information includes location information, the location of the scrap 2 to be selected can be identified. The estimated information on the weight and solubility of the detected scrap can also be estimated using, for example, a trained model, in the same way as obtaining the sorting information.

[0020] <Selective Calculation Unit 5> The selection calculation unit 5 refers to the selection information of each detected scrap obtained by the image analysis unit 4. Then, it executes a process of selecting detected scraps that satisfy preset selection conditions. Here, the preset selection conditions are information about the scraps 2 for selecting the scraps 2 from among the scraps 2 that have been placed and photographed by the imaging device 3. In this embodiment, it is assumed that scrap 2 to be fed into an electric furnace is to be sorted. Therefore, the selection information is set as target component information of the scrap 2 to be fed into the electric furnace. The selection information may be set appropriately depending on the intended use of the scrap 2 to be removed from the scrap group 1 and transported. The selection calculation unit 5 of this embodiment then compares the component estimation information (sorting information) of each detected scrap with the target component information (selection conditions). Based on this comparison, the selection calculation unit 5 selects a detected scrap from the scrap group 1 that will become the target component. The selected detected scrap is called a selected scrap.

[0021] The selection calculation unit 5 of this embodiment, for example, selects detected scraps that are the same as or similar to the target component as the selected scrap. Also, in the case where the target component is determined by detected scraps of multiple regions, the selection calculation unit 5 of this embodiment does the following: That is, the detected scraps of the multiple regions are selected as the selected scrap. Note that, in the case where the target component is determined by multiple selected scraps, for example, the following is done: That is, information that forms a set of the multiple selected scraps is added to the scrap information, etc. Here, if the component estimation information is not component information itself, for example, a process of changing the information to component information is also executed. A case where the component estimation information is not component information itself is when the component estimation information is a type (such as the variety of scrap 2) or a grade. The above-mentioned change process is executed, for example, by preparing a database that stores conversion information between type and component, and referencing the database.

[0022] Furthermore, when the selected scrap is to be directly discharged to the electric furnace, the discharge amount (feed amount) is also important. In this case, a target weight may be set as a selection condition in addition to the target composition information. When a target weight is added to the selection conditions, the selection calculation unit 5 executes a sorting process for the selected selected scrap so that the estimated weight of the selected selected scrap becomes the target weight. Furthermore, when a target weight is added to the selection conditions, it is advisable to set estimated information on the weight of the detected scrap as scrap information.

[0023] <Transfer device 6> The transfer device 6 is a device that takes out and transfers the scrap 2 from the area where the selected scrap selected by the selection calculation unit 5 is located. The position of the selected scrap can be determined by referring to the position information contained in the scrap information. As shown in FIG. 1, the transfer device 6 of this embodiment includes a transfer control unit 6A and a transfer actuator 6B.

[0024] [Transfer actuator 6B] The transfer actuator 6B is an actuator for removing and transporting scrap 2 from the scrap pile 1 that has been placed thereon. The transfer actuator 6B may be any driving device capable of selectively picking up scrap 2. Examples of the transfer actuator 6B include a robot arm alone, a combination of a robot arm and a belt conveyor, or a crane device. However, the transfer actuator 6B is preferably a transfer actuator 6B that can lift the scrap 2 and transport it to a predetermined location. Examples of such a transfer actuator 6B include a lifting magnet (hereinafter also referred to as a lifting magnet), a grapple, and a shovel. Furthermore, a weight measuring device 6C for measuring the weight of the scrap 2 to be transferred to the transfer actuator 6B may be provided (see FIG. 4(b)).

[0025] [Transport control unit 6A] The transfer control unit 6A acquires the position information of the selected scrap selected by the selection calculation unit 5. Then, it controls the drive of the transfer actuator 6B to remove and transport the scrap 2 from the area where the selected scrap is located. Furthermore, when the weight measuring device 6C is provided on the transfer actuator 6B, the transfer amount of the scrap 2 can be adjusted according to the weight of the scrap 2 to be transferred.

[0026] [Example of transfer process] In the example described below, a case where a lift magnet is used as the transfer actuator 6B will be described. The LifMag is capable of adjusting the lifting force of the scrap 2. Methods for adjusting the lifting force include adjusting the strength of the magnetic force of the LifMag, adjusting the range of the magnetic force, adjusting the distance between the LifMag and the scrap 2, and adjusting the contact area between the LifMag and the scrap 2.

[0027] Furthermore, when adjusting the distance, the distance between the lift magnet and the scrap 2 is measured from the measurement results of sensors such as a distance measuring sensor and an image captured by a camera. Then, the distance can be adjusted from the measurement information. Also, with the magnetic force off, bring the lift magnet and the scrap 2 into contact once, and based on this state, lift the lift magnet. By doing this, it is also possible to adjust the distance between the lift magnet and the scrap 2. Also, in this embodiment, as shown in FIG. 4(b), assume that a weight measuring device 6C is attached to the lift magnet. The weight measuring device 6C is, for example, a load cell or the like, and it is possible to measure the actual weight of the scrap 2 lifted by the lift magnet.

[0028] (Operation and others) Hereinafter, as an example of a method for transporting the scrap 2 using the above-described scrap transport device, a case where the transport method is applied to a method for discharging the scrap 2 will be described as an example. The following example is a case where when the scrap 2 is charged into a furnace (such as an electric furnace or a converter), the scrap 2 of the target component is automatically collected using the scrap transport device. The scraps 2 in the scrap group 1 have various varieties (types) and grades. Therefore, their respective component ratios are different. In order to produce iron with an appropriate component ratio in the furnace, it is necessary to charge the furnace with appropriate types and weights of the scrap 2. In this example, the sorting information is used as the component estimation information of the scrap 2, and the selection conditions are used as the target component information.

[0029] <Processing flow> The main flow of the discharging method includes the following Steps 1 to 4 as shown in FIG. 3.

[0030] <Step 1 (Imaging step)> In Step 1, the imaging device 3 images the scrap group 1. In Step 1, the imaging device 3 may image the entire image of the scrap group 1, or the scrap group 1 may be divided into a plurality of regions (see FIG. 2) and images may be taken for each region.

[0031] <Step two (Estimation step)> In each of one or more regions located in the scrap group 1, a collection of one or more scraps 2 existing in each region is defined as a detected scrap. In Step2, the captured image is analyzed to estimate scrap information including the component estimation information (sorting information) of the scrap 2 that identifies the position of each detected scrap and the content of the scraps 2 constituting the detected scrap. In Step2, the component estimation information of the scrap 2 in the scrap group 1 is obtained using the learned model. Also, in addition to the component estimation information of the scrap 2, the estimated weight information of the scrap 2 is obtained.

[0032] Here, it is preferable to estimate the component estimation information (and the estimated weight information) of the scrap 2 by dividing the scrap group 1 into a plurality of regions and estimating each region separately. For example, when the entire image of the scrap group 1 is captured in Step1, the image may be divided into a plurality of regions and the component estimation information (and the estimated weight information) of the scrap 2 may be estimated for each region. When the scrap group 1 is captured for each of a plurality of regions in Step1, the component estimation information (and the estimated weight information) of the scrap 2 may be estimated for each captured image. As shown in FIG. 2, the plurality of regions may be partitioned, for example, in a grid pattern.

[0033] <Step3 (selection step)> In Step3, based on the estimated component estimation information (sorting information) of the scrap 2, detected scraps that satisfy the target component information (pre-set selection conditions) are selected from the scrap group 1. In Step3, based on the estimated component estimation information of the scrap 2 for each region, regions are selected so as to be the target component. Specifically, the ratio of the component (target component) desired in the operation of the furnace is input as an input value to the selection calculation unit 5. Then, it is calculated which scrap 2 (which region's scrap 2) in the captured scrap group 1 should be charged into the furnace (see scraps A and B in FIG. 4(a)). At this time, the target weight of the target component is also input as an input value to the selection calculation unit 5 together with the target component, and the scrap 2 in the scrap group 1 is selected so that the target component becomes the target weight.

[0034] <Step 4 (Dispensing Step)> In Step 4, all or part of the selected detected scraps are taken out and dispensed (transported). In Step 4, Scrap 2 in the area selected in Step 3 is lifted by a lifter magnet and transported. At this time, it is preferable to lift only Scrap 2 in the range (range where components are estimated) shown in the captured image by adjusting the lifting force (such as magnetic force) of the lifter magnet. That is, it is preferable to lift only Scrap 2 on the surface of Scrap Group 1 visible in the captured image.

[0035] Also, when there are a plurality of areas selected in Step 3, Scrap 2 in the plurality of areas is lifted and transported respectively. For example, when the plurality of selected areas are far apart, as shown in Fig. 4(b), first, Scrap 2 (Scrap A) in one area is lifted by a lifter magnet. Then, as shown in Fig. 4(c), the lifter magnet is moved to another area to lift Scrap 2 (Scrap B) in the other area. Note that when lifting Scrap 2 in one area with a lifter magnet and moving it to another area, the situation of Scrap Group 1 may change (the mountains may collapse). Therefore, as shown in Fig. 5(c), it may be configured to perform Steps 1 to 3 again to obtain component estimation information of Scrap 2 in other areas.

[0036] On the other hand, when the plurality of selected areas are close to each other, as shown in Figs. 6 and 7, it is also possible to continuously lift Scrap 2 in the plurality of areas while lowering the lifter magnet and moving it between the areas of Scrap Group 1. In that case, as shown in Fig. 8, for example, while moving the lifter magnet, the distance of the lifter magnet is gradually brought closer to Scrap 2. Or, the magnetic force of the lifter magnet is gradually increased. Thereby, Scrap 2 in the plurality of areas can be continuously lifted.

[0037] In Step 4, the weight of the scrap 2 lifted by the lift mag is measured by the weight measuring device 6C attached to the lift mag. Then, the scrap 2 may be lifted in each region by the estimated weight (target weight) acquired by the weight estimation unit 4C. Specifically, the actual weight information of the scrap 2 lifted by the lift mag is compared with the estimated weight (target weight). If the comparison reveals that the actual weight is less than the estimated weight (target weight), it is possible that not all of the scrap 2 within the range whose composition has been estimated has been lifted. In this case, control may be performed, such as lifting additional scrap 2 in that region. On the other hand, if the actual weight is greater than the target weight, it is possible that scrap 2 outside the range whose composition has been estimated has also been lifted. In this case, control may be performed to weaken the lifting force (magnetic force) of the lift mag and drop some of the lifted scrap 2. Alternatively, control may be performed, such as dropping all of the scrap 2 and then lifting it again.

[0038] (Program 20) The above-described process of selecting scraps 2 from the scrap group 1 and instructing the transportation of the selected scraps 2 may be configured as a program 20 executed by a computer, as shown in Fig. 9. The program 20 may also be stored in a computer-readable recording medium 21. As shown in FIG. 9, the program 20 of this embodiment includes an acquisition step 20A, an estimation step 20B, a selection step 20C, and a transfer instruction step 20D.

[0039] The acquisition step 20A executes a process of acquiring a captured image of a scrap group 1 made up of a plurality of scraps 2. In addition, a collection of one or more scraps 2 present in each of one or more areas located in the scrap group 1 is defined as a detected scrap. The estimation step 20B executes a process of estimating scrap information including selection information that identifies the position of each detected scrap and the contents of the scraps 2 that make up the detected scraps, by image analysis of the acquired captured image.

[0040] The selection step 20C refers to the estimated selection information and executes a process of selecting detected scraps that satisfy a preset selection condition. The transfer instruction step 20D sends a command to the transfer device 6 to take out and transport the selected detected scrap, and executes a process of taking out and transporting the selected detected scrap via the transfer device 6. Such a program 20 is stored in a storage unit 21 (recording medium 21) that constitutes a recording medium such as a hard disk, and is called and executed by a CPU 22 of a computer.

[0041] (Effects and others) According to this embodiment, before removing the scrap 2 to be transported, sorting information is acquired for each detected scrap that constitutes the scrap group 1. Then, based on the sorting information, the scrap 2 to be transported is selected, and the selected scrap 2 is then transported. Therefore, according to this embodiment, it is possible to take out and transport only the scrap 2 that mainly satisfies the selection conditions without analyzing the components of the scrap 2 after picking it up. Furthermore, according to this embodiment, for example, when assuming discharge to an electric furnace, the selection conditions can be set to a target composition, and scrap 2 having the target composition or a composition close to that target composition can be selectively transported.

[0042] (others) The present disclosure may also have the following configuration. (1) Disclosure 1 is a scrap conveying device that selects and conveys scrap from a scrap group consisting of multiple scraps, an imaging device that images the scrap group; an image analysis unit that, when a collection of one or more scraps present in each of one or more areas located in the scrap group is defined as a detected scrap, analyzes the captured image captured by the imaging device and estimates scrap information for each detected scrap, including selection information that identifies the contents of the scraps that make up the detected scrap; a selection calculation unit that refers to the estimated selection information and selects detected scrap that satisfies a preset selection condition; a transfer device that removes and transfers scrap from the area where the detected scrap selected by the selection calculation unit is located; A scrap transport device comprising: (2) Disclosure 2 states that the sorting information is component estimation information that can estimate the components of the scrap, The selection condition is the target component information. (3) Disclosure 3 includes, as the component estimation information, one or more pieces of information selected from information on the components contained in the detected scrap, the type of scrap that constitutes the detected scrap, and the grade of the scrap that constitutes the detected scrap. (4) Disclosure 4 states that the selection calculation unit selects one or more detected scraps based on the component estimation information of each detected scrap so as to approach the target component identified by the target component information. (5) Disclosure 5 relates to the image analysis unit including a weight estimation unit that estimates information on the weight of the detected scrap as one of the scrap information, separate from the sorting information; One of the selection conditions set above is a target weight. (6) In Disclosure 6, the transfer device is configured to remove scrap by lifting a target scrap from a plurality of scraps. (7) Disclosure 7 has a weight measuring device for measuring the weight of the lifted scrap or each group of scrap, The transfer device adjusts the lifting force of the scrap based on the measurement value of the weight measuring device so that the weight of the scrap to be transferred reaches a target weight. (8) In disclosure 8, the scrap transported by the transport device is scrap to be fed into the electric furnace. (9) Disclosure 9 includes information on the solubility of scrap as the sorting information. (10) Disclosure 10 is a scrap conveying method for selecting and conveying scrap from a scrap group consisting of a plurality of scraps, The scrap pile is photographed, When a collection of one or more scraps present in each of one or more areas located in the scrap group is determined as a detected scrap, the captured image is analyzed to estimate scrap information including selection information that identifies the contents of the scraps that make up each detected scrap, By referring to the estimated sorting information, the detected scrap that satisfies the preset selection conditions is selected; removing and transporting all or a part of the selected detected scrap; Scrap transport method. (11) Disclosure 11 includes an acquisition step of acquiring a captured image of a scrap group consisting of a plurality of scraps; an estimation step of determining a collection of one or more scraps present in each of one or more areas located in the scrap group as detected scraps, and estimating scrap information including selection information that identifies the contents of the scraps that make up each detected scrap through image analysis processing of the acquired captured image; a selection step of selecting detected scrap that satisfies a preset selection condition by referring to the estimated selection information; A program that causes a computer to execute the following. (12) A program according to the present disclosure is stored in the program. A computer-readable recording medium.

[0043] The entire contents of Japanese Patent Application No. 2023-189309 (filed November 6, 2023), from which this application claims priority, are incorporated herein by reference. While the present invention has been described with reference to a limited number of embodiments, the scope of the invention is not limited thereto, and modifications of each embodiment based on the above disclosure would be obvious to those skilled in the art. [Explanation of symbols]

[0044] 1. Scrap 2. Scrap 3. Imaging device 4. Image analysis section 4A Image acquisition unit 4B Scrap Information Estimation Department 4C Weight estimation section 5. Selection and calculation unit 6 Transfer device 6A Transfer control section 6B Transfer actuator 6C Weight measuring device 20 Programs 20A Acquisition Steps 20B Estimation Steps 20C Selection Step 20D Transfer Instruction Step 21 Storage medium (storage unit)

Claims

1. A scrap conveying device that selects and conveys scrap from a scrap group consisting of a plurality of scraps, an imaging device that images the scrap group; an image analysis unit that, when a collection of one or more scraps present in each of one or more areas located in the scrap group is defined as a detected scrap, analyzes the captured image captured by the imaging device and estimates scrap information for each detected scrap, including selection information that identifies the contents of the scraps that make up the detected scrap; a selection calculation unit that refers to the estimated selection information and selects detected scrap that satisfies a preset selection condition; a transfer device that removes and transfers scrap from the area where the detected scrap selected by the selection calculation unit is located; Equipped with The image analysis unit includes a weight estimation unit that estimates information on the weight of the detected scrap as one piece of the scrap information, separate from the sorting information, One of the selection conditions set above is that the product has a target weight. Scrap transport device.

2. The sorting information is component estimation information that can estimate the components of the scrap, The selection condition is target component information.

2. A scrap transport device according to claim 1.

3. The component estimation information includes one or more pieces of information selected from information on components contained in the detected scrap, types of scrap constituting the detected scrap, and grades of scrap constituting the detected scrap; 3. A scrap transport device according to claim 2.

4. The selection calculation unit selects one or more detected scraps based on the component estimation information of each detected scrap so that the detected scraps become or approach the target component specified by the target component information.

3. A scrap transport device according to claim 2.

5. The transfer device is configured to remove the scrap by lifting the target scrap from the plurality of scraps.

2. A scrap transport device according to claim 1.

6. a weight measuring device for measuring the weight of the lifted scrap or each group of scrap; the transfer device adjusts the lifting force of the scrap based on the measurement value of the weight measuring device so that the weight of the scrap to be transferred reaches a target weight.

6. A scrap transport device according to claim 5.

7. The scrap transported by the transport device is scrap to be charged into an electric furnace.

3. A scrap transport device according to claim 2.

8. The sorting information includes information on the solubility of the scrap.

8. A scrap transport device according to claim 7.

9. A scrap transport method for selecting and transporting scrap from a scrap group consisting of multiple scraps using a scrap transport device described in any one of claims 1 to 8, comprising: The scrap pile is photographed, When a collection of one or more scraps present in each of one or more areas located in the scrap group is determined as a detected scrap, the captured image is analyzed to estimate scrap information including selection information that identifies the contents of the scraps that make up each detected scrap, By referring to the estimated sorting information, the detected scrap that satisfies the preset selection conditions is selected; removing and transporting all or a part of the selected detected scrap; Scrap transport method.

10. A program for selecting scrap from a scrap group consisting of multiple scraps, which is executed when selecting and transporting scrap using the scrap transport device described in any one of claims 1 to 8, an acquiring step of acquiring an image of a scrap group consisting of a plurality of scraps; an estimation step of determining a collection of one or more scraps present in each of one or more areas located in the scrap group as detected scraps, and estimating scrap information including selection information that identifies the contents of the scraps that make up each detected scrap through image analysis processing of the acquired captured image; a selection step of selecting detected scrap that satisfies a preset selection condition by referring to the estimated selection information; A program that causes a computer to execute the following.

11. A program according to claim 10 is stored in the A computer-readable recording medium.

Images