Scrap iron grading method and device, electronic equipment and media

Abstract

To provide an iron scrap grading method and apparatus, or the like.SOLUTION: A method comprises acquiring a plurality of pieces of single layer iron scrap data. Each piece of single layer iron scrap data includes image data and point group data acquired by collecting data on a target area if iron scrap material in the corresponding layer becomes a surface layer iron scrap material of the iron scrap heap during an unloading process. The method also comprises determining a pile density of the iron scrap heap; for the iron scrap material of each layer of the iron scrap heap, performing image segmentation processing of the single layer iron scrap data corresponding to the iron scrap material of the layer, so as to obtain an image segmentation result for the iron scrap material of the layer, and determining a single layer weight of the iron scrap material based on a single layer volume and a pile density, determining a single layer grading result of the iron scrap material based on the image segmentation result and the single layer weight, and determining an overall grading result for the iron scrap heap based on the plurality of single layer grading results corresponding to a multi-layer iron scrap material.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method, apparatus, electronic device, computer-readable storage medium and computer program product for grading ferrous scrap. [Background technology]

[0002] In response to the country's demand for a green economy and to expedite the realization of carbon-peak and carbon-neutral goals, many steel mills, both domestic and international, are reducing their reliance on iron ore and using scrap steel as an important raw material for steelmaking. Steel mills must purchase large quantities of scrap steel each year, which generally comes in a vast variety of materials, complex shapes, and sizes. Many scrap steels contain impurities such as soil and oil. Traditionally, steel mills rely on visual inspection to grade the purchased scrap steel. After the scrap steel is collected, stored, and transshipped, the grading results are manually entered and uploaded into ERP or MES systems. With the continuous advancement of technology and the continuous development of machine vision, deep learning-based scrap steel grading methods are also beginning to play a role in the field of scrap steel grading.

[0003] In the prior art, generally, images are collected for the scrap iron material in each layer of the scrap iron heap during unloading, and the collected images are subjected to image segmentation processing to identify the number or area of ​​steel materials contained in the scrap iron in each layer, thereby obtaining the classification level of the scrap iron in that layer. Summary of the Invention

[0004] The present disclosure provides methods, apparatus, electronic devices, computer readable storage media and computer program products for grading ferrous scrap.

[0005] According to one aspect of the present disclosure, there is provided a method for grading iron scrap, comprising: acquiring a plurality of single-layer iron scrap data, wherein the plurality of single-layer iron scrap data and multiple layers of iron scrap material contained in an iron scrap heap correspond one-to-one, the iron scrap heap is within a target area, and each single-layer iron scrap data includes image data and point cloud data obtained by collecting data for the target area when the iron scrap material of the corresponding layer becomes a surface layer iron scrap material of the iron scrap heap during an unloading process; determining a pile density of the iron scrap heap; and for each layer of iron scrap material of the iron scrap heap, performing image segmentation for the iron scrap material of the layer. performing an image segmentation process on image data of the single-layer iron scrap data corresponding to the iron scrap heap material, wherein the image segmentation result indicates steel information and non-steel information of the iron scrap material of the layer; determining a single-layer volume of the iron scrap material of the layer based on point cloud data of the single-layer iron scrap data corresponding to the iron scrap material of the layer; determining a single-layer weight of the iron scrap material of the layer based on the single-layer volume and the pile density; determining a single-layer grading result of the iron scrap material of the layer based on the image segmentation result and the single-layer weight; and determining an overall grading result of the iron scrap heap based on at least one single-layer grading result of a plurality of single-layer grading results corresponding to the multi-layer iron scrap material.

[0006] According to another aspect of the present disclosure, there is provided a grading device for iron scrap, the grading device including: a first acquisition module configured to acquire a plurality of single-layer iron scrap data, wherein the plurality of single-layer iron scrap data correspond one-to-one to multiple layers of iron scrap material contained in an iron scrap heap, the iron scrap heap being within a target area, and each single-layer iron scrap data includes image data and point cloud data acquired by performing collection on the target area when the iron scrap material of a corresponding layer becomes a surface layer iron scrap material of the iron scrap heap during an unloading process; a determination module configured to determine a pile density of the iron scrap heap; and a determination module configured to, for each layer of iron scrap material of the iron scrap heap, perform image segmentation for the iron scrap material of the layer to obtain an image segmentation result for the iron scrap material of the layer. The method includes a processing module configured to: perform image segmentation processing on image data of the single-layer iron scrap data corresponding to scrap material, wherein the image segmentation result indicates steel information and non-steel information of the iron scrap material of the layer; determine a single-layer volume of the iron scrap material of the layer based on point cloud data of the single-layer iron scrap data corresponding to the iron scrap material of the layer; determine a single-layer weight of the iron scrap material of the layer based on the single-layer volume and the pile density; and determine a single-layer grading result of the iron scrap material of the layer based on the image segmentation result and the single-layer weight; and determine an overall grading result of the iron scrap heap based on at least one single-layer grading result of a plurality of single-layer grading results corresponding to the multi-layer iron scrap material.

[0007] According to another aspect of the present disclosure, there is provided an electronic device including at least one processor and a memory communicatively coupled to the at least one processor, the memory storing instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the method described above.

[0008] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium having stored thereon computer instructions for causing a computer to perform the above-described method.

[0009] According to another aspect of the present disclosure, there is provided a computer program product including a computer program which, when executed by a processor, implements the above-mentioned method.

[0010] According to one or more embodiments of the present disclosure, when unloading from a scrap iron heap, image data and point cloud data of a single layer of scrap iron material in the scrap iron heap are collected layer by layer and processed to determine volume information of the single layer of scrap iron material based on visual inspection and the point cloud data, and then determine weight information of the scrap iron material in that layer based on the volume information and the pile density of the scrap iron heap. Based on this, the weight information obtained based on the point cloud data can be supplemented when grading the single layer of scrap iron material, thereby improving the accuracy of the grading results.

[0011] It should be understood that the contents described in this section are not intended to identify key or important features of the embodiments of the present disclosure, and are not intended to limit the scope of protection of the present disclosure. Other features of the present disclosure will be easily understood from the following description. [Brief explanation of the drawings]

[0012] The drawings illustrate examples in examples, constitute a part of the specification, and together with the written description serve to explain exemplary embodiments of the examples. The illustrated examples are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar, but not necessarily identical, elements. [Figure 1] FIG. 1 is a schematic diagram illustrating an exemplary system capable of implementing the methods described herein, according to an embodiment of the present disclosure. [Figure 2]1 is an exemplary flow chart illustrating a method for grading ferrous scrap according to an embodiment of the present disclosure. [Figure 3] 1 is an exemplary flow chart illustrating portions of a method for grading ferrous scrap according to an embodiment of the present disclosure. [Figure 4] 4 is an exemplary flowchart illustrating another portion of a method for grading ferrous scrap according to an embodiment of the present disclosure. [Figure 5] 4 is an exemplary flowchart illustrating yet another portion of a method for grading ferrous scrap according to an embodiment of the present disclosure. [Figure 6] 4 is an exemplary flowchart illustrating yet another portion of a method for grading ferrous scrap according to an embodiment of the present disclosure. [Figure 7] 1 is a block diagram illustrating a configuration of a scrap iron grading device according to an embodiment of the present disclosure. [Figure 8] FIG. 1 is a block diagram illustrating an exemplary electronic device that can be used to implement embodiments of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013]

[0023] The following description will be made in conjunction with the drawings to illustrate exemplary embodiments of the present disclosure. Various details of the embodiments of the present disclosure included therein are intended to facilitate understanding and should be considered merely illustrative. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, the following description omits descriptions of known functions and structures.

[0014] In this disclosure, unless otherwise specified, the use of terms such as "first," "second," etc. to describe various elements is not intended to limit the location, timing, or importance of these elements. Such terms are used only to distinguish one element from another. In some instances, a first element and a second element may refer to the same instance of the element, or in some cases, may refer to different instances based on the context.

[0015] The terms used in the description of the various examples of the present disclosure are intended only to describe particular examples and are not intended to be limiting. Unless the context clearly indicates otherwise, unless a specific number of elements is not limited, the element may be one or more. Furthermore, as used in this disclosure, the term "and / or" covers any and all possible combinations of the listed items.

[0016] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[0017] FIG. 1 is a schematic diagram illustrating an example system 100 capable of implementing the methods described herein, according to an example embodiment.

[0018] Referring to FIG. 1, the system 100 includes a client device 110, a server 120, and a network 130 communicatively coupling the client device 110 to the server 120.

[0019] The client device 110 includes a display 114 and a client application (APP) 112 that can be displayed via the display 114. The client application 112 may be an application that requires downloading and installation before execution, or a lightweight application such as a mini-APP (lite app). If the client application 112 is an application that requires downloading and installation before execution, the client application 112 may be pre-installed on the client device 110 and activated. If the client application 112 is a mini-APP, the user 102 can directly execute the client application 112 without installing it on the client device 110 by searching for the client application 112 (e.g., by name) in a host application or by scanning a graphic code (e.g., a barcode, QR code, etc.) of the client application 112. In some embodiments, the client device 110 may be any type of mobile computing device, including a mobile computer, a mobile phone, a wearable computing device (e.g., a head-mounted device such as a smart watch or smart glasses), or another type of mobile device. In some embodiments, client device 110 may alternatively be a stationary computing device, such as a desktop, a server computer, or other type of stationary computing device.

[0020] Server 120 is typically a server deployed by an Internet Service Provider (ISP) or Internet Content Provider (ICP). Server 120 can represent a single server, a cluster of servers, a distributed system, or a cloud server providing underlying cloud services (e.g., cloud database, cloud computing, cloud storage, cloud communication, etc.). As can be appreciated, although server 120 is shown in FIG. 1 communicating with only one client device 110, server 120 can provide background services to multiple client devices simultaneously.

[0021] Examples of network 130 include a combination of communication networks, such as a local area network (LAN), a wide area network (WAN), a domain network (PAN), and / or the Internet. Network 130 may be a wired or wireless network. In some embodiments, technologies and / or formats including HyperText Markup Language (HTML), Extensible Markup Language (XML), etc. may be used to process data exchanged over network 130. All or some links may also be encrypted using encryption technologies such as Secure Sockets Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. In some embodiments, custom and / or proprietary data communication technologies may be used to replace or supplement the data communication technologies described above.

[0022] 1 , the client application 112 may be a scrap grading application that can provide various functions based on the grading of the scrap, such as obtaining and displaying collected images of the scrap and grading results of the scrap. Correspondingly, the server 120 may be a server used in conjunction with the scrap grading application. The server 120 may provide an online scrap grading service to the client application 112 running on the client device 110. Alternatively, the server 120 may provide a local scrap grading service to the client application 112 running on the client device 110.

[0023] The system 100 of FIG. 1 can be configured and operated in a variety of ways to accommodate the various methods and apparatus described in accordance with this disclosure.

[0024] In the related art, generally, images are collected for each layer of scrap iron material in a scrap iron heap during unloading, and the collected images are subjected to image segmentation processing to identify the number or area of ​​steel contained in the scrap iron in each layer, thereby obtaining the classification level of the scrap iron in that layer. However, this grading method based solely on visual image inspection tends to be inaccurate.

[0025] 2 is an exemplary flowchart illustrating a method 200 for grading ferrous scrap according to an embodiment of the present disclosure. A client or server such as that shown in FIG. 1 may be utilized to implement the method 200 in FIG.

[0026] As shown in FIG. 2, an embodiment of the present disclosure provides a method 200 for grading iron scrap, which includes the following steps 210 to 240.

[0027] In step 210, a plurality of single-layer iron scrap data is acquired. The plurality of single-layer iron scrap data and the multi-layer iron scrap materials contained in the iron scrap heap correspond one-to-one, the iron scrap heap is within the target area, and each single-layer iron scrap data includes image data and point cloud data obtained by collecting data for the target area when the iron scrap materials in the corresponding layer become the surface layer iron scrap materials of the iron scrap heap during the unloading process.

[0028] In some embodiments, the target area may be the unloading area where the scrap steel transporter is parked in the steel mill, or may be all or part of the area within the loading platform of the scrap steel transporter. In an example, after calibrating the camera coordinate system and the world coordinate system, the loading platform of the transporter within the camera's field of view may be equally divided into multiple rectangular areas, and one or more of these rectangular areas may be the target area.

[0029] In steel mills, electromagnet suction cups or mobile steel grabbers are generally used to unload the scrap iron heap layer by layer, but the thickness of the scrap iron material that the electromagnet suction cup (or mobile steel grabber) attracts may vary each time. Therefore, in some embodiments, the number of times that the electromagnet suction cup (or mobile steel grabber) attracts the scrap iron material to the target area during unloading can be set as the number of layers of the scrap iron heap in the target area, so that the collected image data and point cloud data can more comprehensively and effectively characterize the scrap iron heap's characteristic information.

[0030] In some embodiments, one or more cameras and 3D scanning equipment can be installed directly above the designated unloading point facing the scrap metal heap to collect the image data and point cloud data described above. In other embodiments, cameras can be installed only to collect image data, and then Neural Radiance Fields (NeRF) technology can be used to generate 3D point cloud data based on the 2D image data, thereby reducing costs.

[0031] In step 220, the pile density of the ferrous scrap heap is determined.

[0032] In some embodiments, for multiple scrap iron heaps on the same hauler, the pile density of the scrap iron heap in the non-target area can be determined, and then that pile density can be directly used as the pile density of the scrap iron heap in the target area.

[0033] In some alternative embodiments, pile density can be calculated directly for the current scrap ferrous heap. In this regard, Figure 3 is an exemplary flow chart illustrating portions of a method for grading scrap ferrous metal according to an embodiment of the present disclosure.

[0034] As shown in FIG. 3, in some embodiments, step 220 includes the following steps 310 to 340.

[0035] In step 310, the total weight of the ferrous scrap heap is obtained.

[0036] In some embodiments, gross weight data and tare weight data measured through a weighbridge when a transport vehicle enters the plant and when it leaves the plant after unloading is complete can be obtained from a weighbridge weighing system at a steel mill, thereby obtaining a net weight of all steel scrap based on the difference between the gross weight and the tare weight.

[0037] In one example, the target area includes all areas within the loading bay, and the net weight mentioned above is the total weight of the iron scrap heap.

[0038] In another example, the loading platform of the transport vehicle may be divided into two identical rectangular areas, and the target area may correspond to one of the rectangular areas. In this case, the total weight of the scrap iron heap may be half the net weight of all the scrap iron. It should be understood that in this case, the total weight of the scrap iron heap may vary depending on the size of the different locations corresponding to the target area.

[0039] In step 320, a single layer volume of the ferrous scrap material for each layer in the multi-layer ferrous scrap material is determined based on the plurality of single layer ferrous scrap data.

[0040] FIG. 4 is an exemplary flow chart illustrating another portion of a method for grading ferrous scrap according to an embodiment of the present disclosure.

[0041] 4, in some embodiments, the target area is within the bed of a truck, and the scrap iron heap includes, from bottom to top, first to nth layers of scrap iron material, where n is an integer greater than or equal to 1. In such a case, step 320 can include the following steps 410 to 430.

[0042] In step 410, the actual height of the vehicle from the lowest point of the platform of the transport vehicle to the ground, the actual length of the area and the actual width of the area of ​​the target area are obtained.

[0043] In step 420, for a first layer of scrap iron material in a scrap iron heap, in one embodiment, step 420 includes steps 421 to 425. In step 421, a first point cloud height from the highest point of the scrap iron material in the first layer to the ground and a first point cloud width of a target area are determined based on point cloud data in the single-layer scrap iron data corresponding to the scrap iron material in the first layer. In step 422, a dimension conversion rate is determined based on the ratio of the actual width of the area to the first point cloud width. In step 423, a first actual height from the highest point of the scrap iron material in the first layer to the ground is determined based on the dimension conversion rate and the first point cloud height. In step 424, an actual height of the first layer of scrap iron material in the first layer is determined based on the difference between the first actual height and the actual height of the vehicle. In step 425, a single-layer volume of the scrap iron material in the first layer is determined based on the actual length of the area, the actual width of the area, and the actual height of the first layer.

[0044] In step 430, for the iron scrap material in the i-th layer of the iron scrap heap, where i is an integer and the value of i ranges from (1, n), in one embodiment, step 430 includes steps 431 to 435. In step 431, a second point cloud height from the highest point of the iron scrap material in the i-th layer to the ground is determined based on point cloud data of the single layer iron scrap data corresponding to the iron scrap material in the i-th layer. In step 432, a second point cloud height from the highest point of the iron scrap material in the i-th layer to the ground is determined based on point cloud data of the single layer iron scrap data corresponding to the iron scrap material in the (i-1)th layer. Based on the point cloud data, a third point cloud height from the highest point of the (i-1)th layer of scrap iron material to the ground is determined. In step 433, a point cloud height of the i-th layer of scrap iron material is determined based on the second point cloud height and the third point cloud height. In step 434, an actual height of the i-th layer of scrap iron material is determined based on the dimension conversion ratio and the point cloud height of the i-th layer. In step 435, a single layer volume of the i-th layer of scrap iron material is determined based on the actual length of the area, the actual width of the area, and the actual height of the i-th layer.

[0045] In steps 420 and 430, first, a dimensional conversion ratio between the size change in the corresponding point cloud data of the same object and the size change in the actual scene where it is located is determined based on the ratio between the actual width of the target area obtained by measurement and the point cloud width of the target area determined from the point cloud data, so that the dimensional values ​​in the point cloud data of the single-layer ferrous scrap material can be directly converted to the dimensional values ​​in the actual scene based on the dimensional conversion ratio. Based on this, there is no need to measure the actual volume of the single-layer ferrous scrap every time the ferrous scrap is sucked in, and data related to the volume of the ferrous scrap material in that layer can be directly determined based on the point cloud data of the single-layer ferrous scrap material, which effectively improves the efficiency of obtaining the volume of the single-layer ferrous scrap material and reduces the difficulty of implementation.

[0046] In the example, the actual height of the vehicle from the lowest point of the platform to the ground is h, the actual width of the vehicle is w, and the actual length of the vehicle is l. Based on the platform length l, the platform can be equally divided into m areas, and the width of each area is w and the length is

[0047]

number

[0048] For the first layer of steel scrap material in the target area, a first point cloud height from its highest point to the ground based on the point cloud data

[0049]

number

[0050]

number

[0051]

number

[0052] First actual height from the highest point of one layer of scrap iron material to the ground

[0053]

number

[0054]

number

[0055] Actual height of the first layer of scrap steel material

[0056]

number

[0057]

number

[0058] Single layer volume of the first layer of scrap iron material

[0059]

number

[0060]

number

[0061] For the iron scrap material in the i-th layer after the first layer in the target area, a second point cloud height from the highest point to the ground based on the point cloud data

[0062]

number

[0063]

number

[0064]

number

[0065]

number

[0066] Actual height of the i-th layer of steel scrap material

[0067]

number

[0068]

number

[0069]

number

[0070]

number

[0071] Referring back to FIG. 3, in step 330, the total volume of the scrap ferrous heap is determined based on the multiple single layer volumes corresponding to the multi-layered scrap ferrous material.

[0072] In some embodiments, the calculated single layer volumes can be added together to obtain the total volume of the ferrous scrap heap, as follows:

[0073]

number

[0074] In some embodiments, if the total weight of the ferrous scrap heap is T, the equation for pile density is:

[0075]

number

[0076] In some embodiments, the volume and pile density of each layer of scrap iron material in each of the m areas in the loading platform can be calculated, and the density distribution curve of the scrap iron material for one vehicle and the weight distribution ratio information of each layer of scrap iron material can be obtained, which can further grade the scrap iron material for one vehicle.

[0077] Referring back to FIG. 2, in step 230, the following steps 2310 to 2340 are performed for each layer of scrap iron material in the scrap iron heap.

[0078] In step 2310, perform image segmentation processing on the image data of the single-layer iron scrap data corresponding to the iron scrap material of the layer, so as to obtain an image segmentation result for the iron scrap material of the layer, where the image segmentation result indicates steel information and non-steel information of the iron scrap material of the layer.

[0079] FIG. 5 is an exemplary flow chart illustrating another portion of a method for grading ferrous scrap according to an embodiment of the present disclosure.

[0080] As shown in FIG. 5, according to some embodiments, the method 2310 includes the following steps 510 to 530.

[0081] In step 510, a semantic segmentation process is performed on the image data of the single layer iron scrap data for the iron scrap material of the layer to obtain a semantic segmentation result, where the semantic segmentation result indicates the result of performing identification and impurity removal on the entire iron scrap material of the layer.

[0082] In some embodiments, the collected image data includes not only iron scrap material, but also information on non-steel objects such as soil scum, oil stains, and irrelevant areas such as the fenders or base plates of the loading compartment or the background outside the vehicle cabin, so that semantic segmentation processing can be performed on the image data of the single-layer iron scrap material to obtain feature information of each semantic feature area.

[0083] In some embodiments, the above-mentioned semantic segmentation results include, but are not limited to, foreground area information indicating ferrous scrap material, background area information indicating non-ferrous scrap material, impurity removal area information indicating non-steel material, and stacking information indicating stack size and stacking method of ferrous scrap material.

[0084] In some embodiments, the semantic segmentation model includes an image data input module, a feature extraction module, and a predicted classification result output module. In some examples, a deep learning neural network model can be used as the semantic segmentation model. For example, an encoder based on a convolutional neural network, such as VGG-16 or Resnet, can be used as a feature extractor in the backbone network portion of the semantic segmentation model, and an encoder based on a Transformer architecture can also be used as a feature extractor in the backbone network portion of the semantic segmentation model.

[0085] In step 520, an instance segmentation process is performed on the image data of the single layer iron scrap data corresponding to the iron scrap material of the layer, so as to obtain an instance segmentation result, where the instance segmentation result indicates the result of identifying and removing impurities for each steel material of the iron scrap material of the layer.

[0086] In some embodiments, data such as steel thickness, steel type, steel size, number of dangerous goods, number of rejected goods, etc. directly affect the grading results of the ferrous scrap heap, so that an instance segmentation process can be performed on the image data of the single layer ferrous scrap material to obtain the characteristic information of the single layer ferrous scrap material.

[0087] In some embodiments, the example segmentation results described above may include, but are not limited to, information indicating the thickness of the elemental ferrous scrap material, information indicating the type of elemental ferrous scrap, information indicating the dimensions of the elemental ferrous scrap material, and information indicating toxic, hazardous, or other rejected material.

[0088] In some embodiments, the instance segmentation model includes an image data input module, a feature extraction module, and a predicted classification result output module. In some embodiments, the above-mentioned instance segmentation process can be realized using a deep learning-based Mask R-CNN model.

[0089] It should be noted that the specific processing steps of the semantic segmentation model and the instance segmentation model described above are prior art, and therefore, the description thereof will be omitted here.

[0090] In step 530, an image segmentation result is generated based on the semantic segmentation result and the example segmentation result.

[0091] By combining the semantic segmentation results and the example segmentation results, feature extraction and complementary correction of iron scrap materials can be performed simultaneously from both the overall and individual perspectives, thereby enriching the extracted image feature information and effectively improving the accuracy of the final iron scrap grading results.

[0092] Referring back to FIG. 2, in step 2320, a single layer volume of the scrap iron material of the layer is determined based on point cloud data of the single layer scrap iron data corresponding to the scrap iron material of the layer.

[0093] In some embodiments, if the pile density of the scrap iron heap in the non-target area of ​​the same truck is directly taken as the pile density in the target area, the actual volume of the single layer scrap iron material can be determined based on the point cloud data by referring to the above steps 420 to 430, which will not be repeated here.

[0094] In some other embodiments, if the pile density is calculated separately for the scrap iron heap in the target area, the single layer volume of the scrap iron material of that layer obtained during the calculation process can be directly obtained.

[0095] In step 2330, a single layer weight of the scrap ferrous material for the layer is determined based on the single layer volume and pile density.

[0096] In some other embodiments, the single layer weight of the scrap iron material of the layer is determined according to the above examples.

[0097]

number

[0098]

number

[0099] In some embodiments, the above image segmentation results and single-layer weights can be further processed based on a decision tree algorithm or a deep learning neural model to obtain a single-layer ranking result.

[0100] In step 240, an overall grading result for the scrap ferrous heap is determined based on at least one single layer grading result of the plurality of single layer grading results corresponding to the multi-layer ferrous scrap material.

[0101] In some embodiments, statistical processing such as weighting, averaging, maximizing or minimizing the results of at least one single layer grading may be performed to obtain an overall grading result for the ferrous scrap heap.

[0102] According to the embodiments of the present disclosure, by introducing point cloud data of single-layer iron scrap, the pile density and volume information of the iron scrap heap can be determined, and the weight information of the single-layer iron scrap can also be determined, so that the iron scrap can be graded based on the image segmentation result and weight information of the single-layer iron scrap image at the same time, which effectively improves the accuracy of iron scrap grading.

[0103] FIG. 6 is an exemplary flowchart illustrating yet another portion of a method for grading ferrous scrap according to an embodiment of the present disclosure.

[0104] As shown in FIG. 6, according to some embodiments, the above-mentioned method 200, in addition to steps 210 to 240, further includes the following steps 610 to 640.

[0105] In step 610, at least one instance segmentation feature data corresponding to at least one target steel material type is obtained from the image segmentation result.

[0106] In step 620, at least one instance segmentation feature data is topologically transformed one by one to obtain at least one steel member figure corresponding to at least one steel member mapped into a topological space.

[0107] In step 630, at least one steel member geometry is matched with a reference steel member geometry of the target member type in a steel member geometry reference database.

[0108] In step 640, the steel information in the image segmentation result is corrected based on the collation result.

[0109] In some embodiments, feature information corresponding to each steel material can be obtained from the image segmentation result as instance segmentation feature data, which represents a feature pattern A of the corresponding steel material. After the feature pattern A is topologically transformed and mapped to a topological space, a corresponding topological feature pattern B can be obtained. Due to the property of homeomorphism, although they have different morphological features in Euclidean space, the properties of the topological morphological patterns corresponding to the same material type are the same. Therefore, by matching the topological feature pattern B with a priori reference steel pattern of the same material type, correction to the type of material type of the steel material can be achieved, thereby improving the accuracy of the grading of scrap steel materials.

[0110] FIG. 7 is a block diagram illustrating a configuration of a scrap iron grading device according to an embodiment of the present disclosure.

[0111] As shown in FIG. 7, an embodiment of the present disclosure provides a grading apparatus for steel scrap, which includes a first acquiring module 710 , a determining module 720 , a processing module 730 and a grading module 740 .

[0112] The first acquisition module 710 is configured to acquire a plurality of single-layer iron scrap data, where the plurality of single-layer iron scrap data and the multi-layer iron scrap material contained in the iron scrap heap correspond one-to-one, the iron scrap heap is within the target area, and each single-layer iron scrap data includes image data and point cloud data obtained by collecting for the target area when the iron scrap material of the corresponding layer becomes the surface layer iron scrap material of the iron scrap heap during the unloading process.

[0113] The determining module 720 is configured to determine a pile density of the ferrous scrap heap.

[0114] The processing module 730 processes, for each layer of scrap ferrous material in the scrap ferrous heap: performing an image segmentation process on image data of the single-layer iron scrap data corresponding to the iron scrap material of the layer, so as to obtain an image segmentation result for the iron scrap material of the layer, wherein the image segmentation result indicates steel information and non-steel information of the iron scrap material of the layer; determining a single layer volume of the iron scrap material of the layer based on point cloud data of the single layer iron scrap data corresponding to the iron scrap material of the layer; determining a single layer weight of the scrap iron material for the layer based on the single layer volume and the pile density; The method is configured to determine a single layer grading result of the scrap iron material of the layer based on the image segmentation result and the single layer weight.

[0115] The grading module 740 is configured to determine an overall grading result for the ferrous scrap heap based on at least one single layer grading result of the plurality of single layer grading results corresponding to the multi-layer ferrous scrap material.

[0116] Here, the operations of the above-mentioned units 710-740 of the iron scrap grading apparatus 700 are similar to the operations of the above-mentioned steps 210-240, respectively, and therefore will not be described again. In some embodiments, the iron scrap grading apparatus 700 may further include one or more additional modules for performing functions corresponding to the additional steps of the method 200 (e.g., steps 610-640), which are not shown in FIG. 7 for the sake of brevity.

[0117] While particular functionality is discussed above with reference to particular modules, it should be noted that the functionality of each module discussed herein may be divided among multiple modules and / or at least some functionality of multiple modules may be combined into a single module. A particular module that performs an operation, as discussed herein, may include the particular module that performs the operation itself, or alternatively, other components or modules that the particular module calls or otherwise accesses to perform the operation (or that perform the operation in conjunction with the particular module). Thus, a particular module that performs an operation may include the particular module that performs the operation itself and / or other modules that the particular module calls or otherwise accesses to perform the operation.

[0118] It should also be understood that various techniques may be described herein in the general context of software hardware elements or program modules. With respect to each module described in FIG. 7 above, it is possible to implement the modules in hardware or in hardware in combination with software and / or firmware. For example, the modules may be implemented as computer program code / instructions configured to be executed on one or more processors and stored on a computer-readable storage medium. Interchangeably, the modules may be implemented as hardware logic / circuitry. The hardware logic / circuitry may include integrated circuit chips (e.g., processors (e.g., including central processing units (CPUs), microcontrollers, microprocessors, digital signal processors (DSPs), etc.)), memory, one or more communication interfaces, and / or one or more components in other circuits) that may optionally execute received program code and / or include embedded firmware to perform functions.

[0119] According to embodiments of the present disclosure, an electronic device, a readable storage medium, and a computer program product are further provided.

[0120] Referring to FIG. 8 , a block diagram of an electronic device 800 functioning as a server or client of the present disclosure will be described, which is an example of a hardware device applicable to various aspects of the present disclosure. The electronic device may represent various forms of digital electronic computing devices, such as laptop computers, desktop computers, stage computers, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing devices, mobile phones, intelligent phones, wearable devices, and other similar computing devices. The components, their connections, and their functions shown herein are merely exemplary and do not limit the implementation of the present disclosure as described and / or claimed herein.

[0121] 8, the electronic device 800 includes a computing unit 801 that can perform various appropriate operations and processes in accordance with a computer program stored in a read-only memory (ROM) 802 or loaded from a storage unit 808 into a random access memory (RAM) 803. The RAM 803 may also store various programs and data necessary for the operation of the electronic device 800. The computing unit 801, the ROM 802, and the RAM 803 are connected to one another via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0122] Several components of the electronic device 800, including an input unit 806, an output unit 807, a storage unit 808, and a communication unit 809, are connected to the I / O interface 805. The input unit 806 may be any type of device capable of inputting information to the electronic device 800. The input unit 806 may receive input numeric or character information and generate key signal input for user settings and / or function control of the electronic device, including, but not limited to, a mouse, a keyboard, a touchscreen, a trackboard, a trackball, a control lever, a microphone, and / or a remote control. The output unit 807 may be any type of device capable of presenting information, including, but not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 808 may include, but is not limited to, a magnetic disk or an optical disk. The communication unit 809 enables the electronic device 800 to exchange information / data with other devices via a computer network, e.g., the Internet, and / or various telecommunication networks, and may include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, e.g., a Bluetooth device, an 802.11 device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0123] The computing unit 801 may be various general-purpose and / or special-purpose processing components having processing and computing capabilities. Examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, computing units that execute various machine learning network algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs various methods and processes described above, such as method 200. For example, in some embodiments, method 200 may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, some or all of the computer program may be loaded and / or installed into the electronic device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, it may perform one or more steps of the method 200 described above. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the method 200 using any other suitable means (eg, firmware).

[0124] Various embodiments of the systems and techniques described herein may be implemented in digital electronic circuitry systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include being embodied in one or more computer programs that may be executed and / or interpreted by a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor, and may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0125] Program code implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, so that when executed by the processor or controller, the program code performs the functions / operations specified in the flowcharts and / or block diagrams. The program code may be entirely executed by machine, partially executed by machine, partially executed by machine and partially executed on a remote machine as a separate software package, or entirely executed on a remote machine or server.

[0126] In the context of this disclosure, a machine-readable medium may be a tangible medium, including or storing a program for use in or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include an electrical connection with one or more leads, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0127] To provide for user interaction, a computer may implement the systems and techniques described herein and include a display device (e.g., a CRT (Cathode Ray Tube) or LCD (Liquid Crystal Display) monitor) for displaying information to a user, and a keyboard and pointing device (e.g., a mouse or trackball) through which a user may provide input to the computer. Other types of devices may also be used to provide for user interaction, for example, providing feedback to a user in any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback) and receiving input from a user in any form (including sound input, speech input, or tactile input).

[0128] The systems and techniques described herein may be implemented in a computing system including backstage components (e.g., as a data server), middleware components (e.g., as an application server), front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with the system or technique implementation), or any combination of backstage components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communications network). Examples of communications networks include a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0129] The computer system may include a client and a server. The client and the server are generally remote from each other and usually interact via a communication network. The client-server relationship is created by running computer programs on corresponding computers. The server may be a cloud server, a server in a distributed system, or a server combined with a blockchain.

[0130] It should be understood that the various forms of flow described above may be used to rearrange, add, or remove steps, and for example, the steps described in this disclosure may be performed in parallel, sequentially, or in a different order, as long as the technical solutions disclosed in this disclosure can achieve the desired results, and the present disclosure is not limited thereto.

[0131] Although embodiments or examples of the present disclosure have been described with reference to the drawings, it should be understood that the above-described methods, systems, and devices are merely exemplary embodiments or examples, and that the scope of the present invention is not limited by these embodiments or examples, but only by the appended claims and their equivalents. Various elements of the embodiments or examples may be omitted or replaced by their equivalents. Furthermore, steps may be performed in a different order than described in this disclosure. Furthermore, various elements of the embodiments or examples may be combined in various ways. Importantly, as technology evolves, many elements described herein may be replaced by equivalent elements that appear later in this disclosure.

Claims

1. A method for grading iron scrap, which is executed by an iron scrap grading device, comprising: acquiring a plurality of single-layer iron scrap data by a first acquisition module, wherein the plurality of single-layer iron scrap data and multiple layers of iron scrap material contained in an iron scrap heap correspond one-to-one, the iron scrap heap is within a target area, and each single-layer iron scrap data includes image data and point cloud data obtained by collecting data for the target area when the iron scrap material in the corresponding layer becomes the surface layer iron scrap material of the iron scrap heap during unloading; determining a pile density of the ferrous scrap heap with a determination module; for each layer of scrap iron material of said scrap iron heap by a processing module: performing an image segmentation process on image data of the single-layer iron scrap data corresponding to the iron scrap material of the layer to obtain an image segmentation result for the iron scrap material of the layer, the image segmentation result indicating steel material information and non-steel material information of the iron scrap material of the layer; determining a single layer volume of the iron scrap material of the layer based on point cloud data of the single layer iron scrap data corresponding to the iron scrap material of the layer; determining a single layer weight of the scrap iron material for the layer based on the single layer volume and the pile density; determining a single layer grading result of the scrap iron material of the layer based on the image segmentation result and the single layer weight; determining, by a grading module, an overall grading result for the ferrous scrap heap based on at least one single layer grading result among a plurality of single layer grading results corresponding to the multi-layer ferrous scrap material; method.

2. The step of determining the pile density of the ferrous scrap heap comprises: obtaining a total weight of the ferrous scrap heap; determining a single layer volume of the ferrous scrap material for each layer in the multi-layer ferrous scrap material based on the plurality of single layer ferrous scrap data; determining a total volume of the scrap ferrous heap based on a plurality of single layer volumes corresponding to the multi-layered scrap ferrous material; and determining the pile density based on the total weight and the total volume.

3. The target area is located within the loading platform of a transport vehicle, and the scrap iron heap includes, from bottom to top, first to n-th layers of scrap iron material, where n is an integer equal to or greater than 1; The step of determining a single layer volume of the iron scrap material of each layer in the multi-layer iron scrap material based on a plurality of single layer iron scrap data includes: obtaining an actual height of the vehicle from the lowest point of the platform of the transporter to the ground, an actual length of the area, and an actual width of the area of ​​the target area; for the scrap iron material of a first layer of said scrap iron heap, determining a first point cloud height from the highest point of the first layer of iron scrap material to the ground and a first point cloud width of the target area based on point cloud data of the single layer iron scrap data corresponding to the first layer of iron scrap material; determining a size conversion ratio based on a ratio between the actual width of the area and the first point cloud width; determining a first actual height from a highest point of the first layer of scrap iron material to ground level based on the size conversion ratio and the first point cloud height; determining an actual height of the first layer of scrap ferrous material based on a difference between the first actual height and an actual height of the vehicle; determining a single layer volume of the first layer of scrap iron material based on an actual length of the area, an actual width of the area, and an actual height of the first layer; For the iron scrap material in the i-th layer of the iron scrap heap, where i is an integer and the range of values ​​of i is (1, n), determining a second point cloud height from a highest point of the iron scrap material in the i-th layer to the ground based on point cloud data in the single-layer iron scrap data corresponding to the iron scrap material in the i-th layer; determining a third point cloud height from the highest point of the iron scrap material in the i-1th layer to the ground based on point cloud data in the single-layer iron scrap data corresponding to the iron scrap material in the i-1th layer; determining a point cloud height of the i-th layer of the iron scrap material based on the second point cloud height and the third point cloud height; determining an actual height of the ith layer of the iron scrap material based on the size conversion rate and the point cloud height of the ith layer; determining a single layer volume of the ith layer of scrap iron material based on an actual length of the area, an actual width of the area, and an actual height of the ith layer.

4. The step of performing image segmentation processing on image data of the single layer iron scrap data corresponding to the iron scrap material of the layer includes: performing a semantic segmentation process on the image data of the single layer iron scrap data for the iron scrap material of the layer to obtain a semantic segmentation result, the semantic segmentation result indicating the result of performing classification and impurity removal on the entire iron scrap material of the layer; performing an instance segmentation process on image data of the single layer iron scrap data corresponding to the iron scrap material of the layer to obtain an instance segmentation result, the instance segmentation result indicating the result of identifying and removing impurities from each steel material of the iron scrap material of the layer; generating the image segmentation result based on the semantic segmentation result and the example segmentation result.

5. obtaining at least one example segmentation feature data corresponding to at least one target steel material type from the image segmentation result; Topologically transforming the at least one instance segmentation feature data one by one to obtain at least one steel product diagram corresponding to the at least one target steel product type mapped into a topological space; Matching the at least one steel member figure with a reference steel member figure of the target member type in a steel member figure reference database; The method of claim 1 , further comprising the step of correcting steel product information in the image segmentation result based on the matching result.

6. A grading device for iron scrap, the grading device for iron scrap comprising: a first acquisition module configured to acquire a plurality of single-layer iron scrap data, wherein the plurality of single-layer iron scrap data and multiple layers of iron scrap material contained in an iron scrap heap correspond one-to-one, the iron scrap heap is within a target area, and each single-layer iron scrap data includes image data and point cloud data obtained by collecting data for the target area when the iron scrap material in the corresponding layer becomes a surface layer iron scrap material of the iron scrap heap during an unloading process; a determination module configured to determine a pile density of the ferrous scrap heap; For each layer of scrap iron material in the scrap iron heap: performing an image segmentation process on image data of the single-layer iron scrap data corresponding to the iron scrap material of the layer to obtain an image segmentation result for the iron scrap material of the layer, the image segmentation result indicating steel material information and non-steel material information of the iron scrap material of the layer; determining a single layer volume of the iron scrap material of the layer based on point cloud data of the single layer iron scrap data corresponding to the iron scrap material of the layer; determining a single layer weight of the scrap iron material for the layer based on the single layer volume and the pile density; a processing module configured to determine a single layer grading result of the scrap iron material of the layer based on the image segmentation result and the single layer weight; a grading module configured to determine an overall grading result for the ferrous scrap heap based on at least one single layer grading result among a plurality of single layer grading results corresponding to the multi-layer ferrous scrap material; Scrap steel grading equipment.

7. The decision module: an acquisition sub-module configured to acquire a total weight of the ferrous scrap heap; a first determination sub-module configured to determine a single layer volume of the ferrous scrap material of each layer in the multi-layer ferrous scrap material based on the plurality of single layer ferrous scrap data; a second determining sub-module configured to determine a total volume of the scrap ferrous heap based on a plurality of single layer volumes corresponding to the multi-layered scrap ferrous material; and a third determining sub-module configured to determine the pile density based on the total weight and the total volume.

8. The target area is located within the loading platform of a transport vehicle, and the scrap iron heap includes, from bottom to top, first to n-th layers of scrap iron material, where n is an integer greater than or equal to 1; The first decision sub-module: Acquire the actual height of the vehicle from the lowest point of the platform of the transport vehicle to the ground, and the actual length and width of the target area; for the scrap iron material of the first layer of said scrap iron heap, determining a first point cloud height from the highest point of the first layer of iron scrap material to the ground and a first point cloud width of the target area based on point cloud data of the single-layer iron scrap data corresponding to the first layer of iron scrap material; determining a size conversion ratio based on a ratio between the actual width of the area and the first point cloud width; determining a first actual height from a highest point of the first layer of scrap iron material to ground level based on the size conversion ratio and the first point cloud height; determining an actual height of the first layer of scrap ferrous material based on a difference between the first actual height and an actual height of the vehicle; determining a single layer volume of the scrap iron material of the first layer based on the actual length of the area, the actual width of the area, and the actual height of the first layer; For the iron scrap material in the i-th layer of the iron scrap heap, where i is an integer and the range of values ​​of i is (1, n), determining a second point cloud height from a highest point of the iron scrap material in the i-th layer to the ground based on point cloud data in the single-layer iron scrap data corresponding to the iron scrap material in the i-th layer; determining a third point cloud height from the highest point of the iron scrap material in the i-1th layer to the ground based on point cloud data in the single-layer iron scrap data corresponding to the iron scrap material in the i-1th layer; determining a point cloud height of the i-th layer of the iron scrap material based on the second point cloud height and the third point cloud height; determining an actual height of the ith layer of the iron scrap material based on the size conversion rate and the point cloud height of the ith layer; 8. The grading device for iron scrap according to claim 7, configured to determine a single layer volume of the iron scrap material of the ith layer based on an actual length of the area, an actual width of the area, and an actual height of the ith layer.

9. The processing module includes: performing a semantic segmentation process on the image data of the single layer iron scrap data for the iron scrap material of the layer to obtain a semantic segmentation result, the semantic segmentation result indicating the result of performing identification and impurity removal on the entire iron scrap material of the layer; performing an instance segmentation process on image data of the single-layer iron scrap data corresponding to the iron scrap material of the layer to obtain an instance segmentation result, the instance segmentation result indicating the result of identifying and removing impurities from each steel material of the iron scrap material of the layer; The iron scrap grading device according to claim 6 , configured to generate the image segmentation result based on the semantic segmentation result and the example segmentation result.

10. a second acquisition module for acquiring at least one example segmentation feature data corresponding to at least one target steel material type from the image segmentation result; a transformation module for performing topological transformation on the at least one instance segmentation feature data one by one to obtain at least one steel product figure corresponding to the at least one target steel product type mapped into a topological space; a matching module for matching the at least one steel member pattern with a reference steel member pattern of the target member type in a steel member pattern reference database; 7. The iron scrap grading device according to claim 6, further comprising: a correction module for correcting steel information in the image segmentation result based on the matching result.

11. An electronic device, at least one processor; a memory communicatively coupled to the at least one processor; 6. An electronic device, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the method of any one of claims 1 to 5.

12. 6. A non-transitory computer readable storage medium having stored thereon computer instructions that cause a computer to perform the method of any one of claims 1 to 5.

13. A computer program comprising instructions which, when executed by a processor, implement the method of any one of claims 1 to 5.

Images