Automatic steel scrap weight metering method and device, storage medium and electronic equipment

Abstract

The invention provides a scrap steel weight automatic metering method. The method comprises the following steps: continuously acquiring weight information of a weighbridge and images of scrap steel transferred to a steel bucket; if the weight information at the Nth moment and the weight information at the (N-1) th moment change, determining the variable quantity of the weight information and a target scrap steel image corresponding to the variable quantity; inputting the target scrap steel image into a pre-trained scrap steel type identification model to obtain a scrap steel type identification result corresponding to the target scrap steel image output by the scrap steel type identification model, and determining the weight demand value of various types of scrap steel at the Nth moment based on the weight demand value of various types of scrap steel at the (N-1) th moment, the variable quantity and the scrap steel type identification result; and outputting display information for indicating the weight demand values of the various types of waste steel at the Nth moment for an operator to check. According to the method provided by the invention, the function of automatically metering the weight of the scrap steel is realized, the working intensity of field operators can be reduced, and the accuracy of scrap steel metering can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of scrap steel weighing, and more specifically, to a scrap steel weight automatic measuring method, device, storage medium and electronic equipment. Background Art

[0002] For the process of steelmaking with scrap steel, it is first necessary to determine the type and weight of steel to be made. Based on the type and weight, the type of scrap steel required and the weight requirements for each type of scrap steel can be determined. After determining the type of scrap steel required and the weight requirements for each type of scrap steel, use an overhead crane or grab bucket to transfer the already piled scrap steel raw materials to the steel bucket. The overhead crane or grab bucket can only transfer one type of scrap steel each time (because the scrap steel raw materials have been piled and placed, and different types of scrap steel raw materials are kept at a certain distance). When the weight of each type of scrap steel in the steel bucket meets the corresponding weight requirements, that is, the scrap steel in the steel bucket has met the steelmaking requirements, the steel bucket can be sent to the converter. In the actual production process, whether the overhead crane operation or the grab bucket operation method is used, it is difficult for the operator in the main control room to grasp the scrap steel type information during the loading process, and can only rely on manual experience to estimate the weight information based on the number of operations of different types of scrap steel.

[0003] Therefore, the prior art method for measuring the weight of scrap steel in the scrap steel feeding link in the steelmaking process uses manual observation and estimation by operators, which has the problems of high work intensity for on-site operators and low accuracy in scrap steel measurement. Summary of the invention

[0004] The present invention provides a method, device, storage medium and electronic device for automatic weight measurement of scrap steel, so as to solve the problem that the scrap steel weight measurement method in the prior art for the scrap steel feeding link in the steelmaking process adopts the manual observation and estimation method of the operator, which has the problems of high work intensity of on-site operators and low accuracy of scrap steel measurement.

[0005] In a first aspect, the present invention provides a method for automatically measuring the weight of scrap steel, which is applied to a scrap steel recycling system. The scrap steel recycling system includes an automatic measuring device for the weight of scrap steel, an overhead crane, a floor scale and a steel bucket. During the operation of the overhead crane by an operator, only one type of scrap steel can be grabbed and placed in the steel bucket at any one time. The steel bucket is located on the floor scale. The floor scale is used to measure the total weight of the steel bucket and the scrap steel in the steel bucket. The automatic measuring device for the weight of scrap steel is electrically connected to the floor scale. The method includes:

[0006] Continuously collecting weight information of the floor scale and images of scrap steel transferred to the steel bucket;

[0007] If there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the target scrap steel image corresponding to the change amount;

[0008] Input the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial to-be-trained network for training;

[0009] Determine the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type identification result;

[0010] Outputting display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

[0011] Optionally, according to the method for automatic weight measurement of scrap steel provided by the present invention, the continuously collecting the weight information of the weighbridge specifically includes:

[0012] The weight information of the scale is continuously collected through the serial interface and sent to the corresponding Socket interface.

[0013] Optionally, according to the method for automatic weight measurement of scrap steel provided by the present invention, if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the scrap steel image corresponding to the change amount specifically includes:

[0014] If the number of socket interface messages received between the N-1th moment and the Nth moment that are different from the weight information at the N-1th moment and have the same weight information difference exceeds the preset threshold;

[0015] The weight information difference is determined as a change in the weight information between the Nth moment and the N-1th moment, and a target scrap steel image corresponding to the change is determined.

[0016] Optionally, according to the method for automatic weight measurement of scrap steel provided by the present invention, determining the target scrap steel image corresponding to the change amount specifically includes:

[0017] Based on the Nth moment and the preset scrap steel transfer time, determine the scrap steel image collection moment;

[0018] The scrap steel image corresponding to the scrap steel collection moment is selected from the continuously collected scrap steel images as the target scrap steel image.

[0019] Optionally, according to the method for automatic scrap steel weight measurement provided by the present invention, the preset scrap steel transfer time ranges from 0 seconds to 15 seconds.

[0020] Optionally, according to the method for automatic weight measurement of scrap steel provided by the present invention, before constructing the training sample set, the method further comprises:

[0021] Using a camera device to collect images of different types of scrap steel in different environments, an initial scrap steel image set is obtained;

[0022] Each image in the initial scrap steel image set is rotated, scaled and flipped to obtain a plurality of sample scrap steel images.

[0023] Optionally, according to the automatic scrap steel weight measurement method provided by the present invention, the structure of the initial network to be trained is a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a third convolutional layer, a third pooling layer, a fourth convolutional layer and a fully connected layer connected in sequence, wherein the first convolutional layer includes 16 convolutional layers, the second convolutional layer includes 8 convolutional layers, the third convolutional layer includes 4 convolutional layers, and the fourth convolutional layer includes 2 convolutional layers.

[0024] In a second aspect, the present invention provides an automatic weight metering device for scrap steel, which is applied to a scrap steel recycling system. The scrap steel recycling system comprises the automatic weight metering device for scrap steel, an overhead crane, a floor scale and a steel bucket. During the operation of the overhead crane by an operator, only one type of scrap steel can be grabbed and placed in the steel bucket at any one time. The steel bucket is located on the floor scale. The floor scale is used to measure the total weight of the steel bucket and the scrap steel in the steel bucket. The automatic weight metering device for scrap steel is electrically connected to the floor scale. The device comprises:

[0025] A collection unit, specifically used to continuously collect the weight information of the floor scale and the image of the scrap steel transferred to the steel bucket;

[0026] a determination unit, configured to determine, if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, a change amount of the weight information and a target scrap steel image corresponding to the change amount;

[0027] an identification unit, configured to input the target scrap steel image into a pre-trained scrap steel type identification model, and obtain a scrap steel type identification result corresponding to the target scrap steel image output by the scrap steel type identification model, wherein the scrap steel type identification model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial to-be-trained network for training;

[0028] A measuring unit, used for determining the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type identification result;

[0029] The display unit is used to output display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

[0030] In a third aspect, the present invention provides a computer-readable storage medium, in which at least one program code is stored, and the at least one program code is loaded and executed by a processor to implement the operations performed by the automatic scrap steel weight measurement method as described in the first aspect.

[0031] In a fourth aspect, the present invention provides an electronic device, comprising one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the operations performed by the automatic scrap steel weight measurement method as described in the first aspect.

[0032] The method, device, storage medium and electronic device for automatic weight measurement of scrap steel provided by the present invention continuously collect the weight information of the floor scale and the image of the scrap steel transferred to the steel bucket. When the weight information of the floor scale changes, the change in weight information, i.e., the weight of the scrap steel transferred to the steel bucket at the Nth moment, is determined, and the target scrap steel image corresponding to the change in weight, i.e., the image of the scrap steel corresponding to the weight information jump at the Nth moment is determined. The target scrap steel image is input into a pre-trained scrap steel type recognition model to obtain the corresponding scrap steel type recognition result. In this way, the type of scrap steel transferred to the steel bucket at the Nth moment and the weight of the scrap steel type transferred to the steel bucket can be obtained. Based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the weight demand values ​​of various types of scrap steel at the Nth moment can be updated, and the updated weight demand values ​​of various types of scrap steel are output and displayed for the operator to view and learn the remaining weights of various types of scrap steel. The scrap steel type recognition model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained. The automatic scrap steel weight measuring method, device, storage medium and electronic equipment provided by the present invention realize the automatic scrap steel weight measuring function, which can reduce the work intensity of on-site operators and improve the accuracy of scrap steel measuring.

[0033] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is a schematic diagram of a flow chart of a method for automatically measuring the weight of scrap steel provided by an embodiment of the present invention;

[0036] Figure 2 A schematic diagram of the structure of an initial network to be trained provided by an embodiment of the present invention;

[0037] Figure 3 A schematic diagram of the structure of an automatic scrap steel weight measurement system based on CNN image recognition technology and weighing signals provided in an embodiment of the present invention;

[0038] Figure 4 A schematic diagram of the structure of an automatic scrap steel weight measuring device provided by an embodiment of the present invention;

[0039] Figure 5 A schematic diagram of the physical structure of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0040] Example implementations will now be described more fully with reference to the accompanying drawings. However, example implementations can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; on the contrary, these implementations are provided so that the present application will be more comprehensive and complete, and the concepts of the example implementations will be fully conveyed to those skilled in the art. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present invention.

[0041] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0042] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0043] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0044] It should be noted that the "multiple" mentioned in this article refers to two or more. "And / or" describes the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0045] Combine the following Figure 1-Figure 5 The invention describes a method, device, storage medium and electronic device for automatic weight measurement of scrap steel. The method is applied to a scrap steel recycling system, which includes an automatic weight measurement device for scrap steel, an overhead crane, a floor scale and a steel bucket. During the operation of the overhead crane by an operator, only one type of scrap steel can be grabbed and placed in the steel bucket at any one time. The steel bucket is located on the floor scale. The floor scale is used to measure the total weight of the steel bucket and the scrap steel in the steel bucket. The automatic weight measurement device for scrap steel is electrically connected to the floor scale. Figure 1 is a flow chart of a method for automatically measuring the weight of scrap steel provided by an embodiment of the present invention, wherein the executing body is an automatic measuring device for the weight of scrap steel, such as Figure 1 As shown, the method of the present invention comprises the following steps:

[0046] Step 110, continuously collecting the weight information of the floor scale and the image of the scrap steel transferred to the steel bucket.

[0047] Specifically, the floor scale is an electronic scale with communication function, which can measure the weight of the heavy objects placed on the floor scale in real time. Since the floor scale is electrically connected to the automatic scrap weight metering device, the automatic scrap weight metering device can continuously collect the weight information of the electronic scale. For example, the automatic scrap weight metering device can continuously collect the weight information of the electronic scale through the serial interface. In addition, a high-definition camera device is set up at a suitable position near the floor scale to ensure that the image of the scrap steel during the transfer and loading process can be collected. The clearer the collected image, the better the subsequent recognition effect.

[0048] Step 120: If there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determine the change amount of the weight information and the target scrap steel image corresponding to the change amount.

[0049] Specifically, if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, the change in the weight information is determined. Here, the Nth moment and the N-1th moment are explained. Due to the limitation of the operating mode of the overhead crane, the time between the two moments of transferring scrap steel to the steel bucket is relatively large, so a preset time can be set to judge the change in weight information between the two moments. For example, when the preset time is set to 4 seconds, the i-th moment represents the 4×i-th ​​second, the N-th moment represents the 4×N-th second, and the N-1-th moment represents the 4×(N-1)-th second. Then, setting the preset time to 4 seconds is equivalent to judging the change in weight information every 4 seconds. The setting of the preset time is based on the operating mode of the overhead crane and the stable time after the weight jump. If the preset time is too long, the measurement of the scrap steel transferred to the steel bucket may be missed. If the preset time is too short, the weight of the scrap steel transferred to the steel bucket once may still be shaking and changing, and it may be mistakenly judged as the weight of the scrap steel entering the steel bucket multiple times.

[0050] In addition, since there is a camera device installed at a suitable position near the floor scale to continuously collect scrap steel images, when the weight information of the floor scale jumps, after determining the change in weight information between the two previous and subsequent moments, the target scrap steel image of the transported scrap steel that causes the weight information jump can also be determined. The determination method can be to estimate a transport time (the transport time refers to the time between the start time when the scrap steel enters the corresponding range of the floor scale to the time when the scrap steel is put into the steel bucket), and determine the acquisition time of the target scrap steel image as the Nth moment minus the estimated transport time. For example, the Nth moment is the 457th second after the start of the operation, and the estimated transport time is 13 seconds. Then the transported scrap steel that causes the jump in weight information at the Nth moment (i.e., the 457th second) should be captured in the target scrap steel image collected at 457-13=444 seconds. Therefore, the scrap steel image collected at the 444th second is selected from the continuously collected scrap steel images as the target scrap steel image corresponding to the change.

[0051] Step 130, inputting the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of multiple sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained.

[0052] Specifically, the target scrap steel image is input into the pre-trained scrap steel type recognition model to obtain the scrap steel type recognition result corresponding to the output of the scrap steel type recognition model. Therefore, before executing step 130, the scrap steel type recognition model can also be pre-trained. Specifically, the scrap steel type recognition model can be trained in the following way: first, a large number of sample scrap steel images are collected. The collection process can be to collect scrap steel images of different angles, different lighting conditions, different sizes and different types, so as to improve the generalization ability of the trained scrap steel type recognition model, ensure that the number of scrap steel samples of each category is roughly balanced, and avoid the trained scrap steel type recognition model being biased towards outputting a large number of scrap steel categories. The data diversity is increased by means of rotation, scaling, flipping, etc., and the robustness of the model to deformation, lighting changes, etc. is improved. The image is standardized to accelerate the convergence speed of the scrap steel type recognition model and improve the training stability. Then, the corresponding scrap steel type label is determined for each sample scrap steel image, and the scrap steel type label can be manually calibrated. The initial network to be trained is trained based on a large number of sample scrap steel images and corresponding scrap steel type labels, so as to obtain a scrap steel type recognition model. The initial network to be trained may be a single neural network model or a combination of multiple neural network models. The embodiment of the present invention does not specifically limit the type and structure of the initial network to be trained.

[0053] Step 140, determining the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type identification result.

[0054] Specifically, as the overhead crane transfers scrap steel to the steel bucket each time, the weight demand values ​​of various types of scrap steel are constantly changing. For example, at the beginning, 15 tons of type A scrap steel, 10 tons of type B scrap steel, and 8 tons of type C scrap steel are needed. The first time the overhead crane transferred 1 ton of type A scrap steel, then the weight demand values ​​of the remaining types of scrap steel at the corresponding moment were: 14 tons of type A scrap steel, 10 tons of type B scrap steel, and 8 tons of type C scrap steel. The second time the overhead crane transferred 2 tons of type B scrap steel, then the weight demand values ​​of the remaining types of scrap steel at the corresponding moment were: 14 tons of type A scrap steel, 8 tons of type B scrap steel, 8 tons of type C scrap steel, and so on. Therefore, after determining the weight demand values ​​of various types of scrap steel at the N-1th moment, as well as the change amount and the target scrap steel type at the Nth moment, the weight demand values ​​of various types of scrap steel at the Nth moment are equal to the weight demand values ​​of various types of scrap steel at the N-1th moment minus the target scrap steel type of the change amount. For example, after determining that the weight demand values ​​of various types of scrap steel at the N-1th moment are: 14 tons for type A scrap steel, 8 tons for type B scrap steel, and 8 tons for type C scrap steel, and the change amount at the Nth moment is 1.5 tons, and the scrap steel type is type B, then the updated weight demand values ​​of various types of scrap steel (i.e., the weight demand values ​​of various types of scrap steel at the Nth moment) are: 14 tons for type A scrap steel, 6.5 tons for type B scrap steel, and 8 tons for type C scrap steel.

[0055] Step 150: outputting display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

[0056] Specifically, the latest required weight values ​​of various types of scrap steel are displayed for the operator to view, and the operator can make an estimate of the grab weight of various types of scrap steel next time after viewing.

[0057] The method provided by the embodiment of the present invention continuously collects the weight information of the floor scale and the image of the scrap steel transferred to the steel bucket. When the weight information of the floor scale changes, the change in weight information, i.e., the weight of the scrap steel transferred to the steel bucket at the Nth moment, is determined, and the target scrap steel image corresponding to the change in weight, i.e., the image of the scrap steel corresponding to the weight information jump at the Nth moment is determined. The target scrap steel image is input into the pre-trained scrap steel type recognition model to obtain the corresponding scrap steel type recognition result, so that the type of scrap steel transferred to the steel bucket at the Nth moment and the weight of the scrap steel type transferred to the steel bucket can be obtained. Based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the weight demand values ​​of various types of scrap steel at the Nth moment can be updated, and the updated weight demand values ​​of various types of scrap steel are output and displayed for the operator to view and learn the remaining required weights of various types of scrap steel, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of multiple sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained. The automatic scrap steel weight measuring method provided by the present invention realizes the automatic scrap steel weight measuring function, can reduce the work intensity of on-site operators, and improve the scrap steel weight measuring accuracy.

[0058] Based on the above embodiment, in the method, the continuously collecting the weight information of the floor scale specifically includes:

[0059] The weight information of the scale is continuously collected through the serial interface and sent to the corresponding Socket interface.

[0060] Specifically, the floor scale and the automatic scrap steel weight measuring device are electrically connected, that is, the automatic scrap steel weight measuring device continuously collects the weight information of the floor scale through the serial interface, and then sends it to the corresponding Socket interface, wherein the Socket interface is an interface connecting the upper-level application (i.e., the front-end web page program) and the underlying communication structure (i.e., the background operating system). After the automatic scrap steel weight measuring device collects the weight information of the floor scale through the serial interface, it needs to send it to the Socket interface corresponding to the automatic scrap steel weight measuring program.

[0061] Based on any of the above embodiments, in the method, if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the scrap steel image corresponding to the change amount specifically includes:

[0062] If the number of socket interface messages received between the N-1th moment and the Nth moment that are different from the weight information at the N-1th moment and have the same weight information difference exceeds the preset threshold;

[0063] The difference is determined as a change in weight information between the Nth moment and the N-1th moment, and a target scrap steel image corresponding to the change is determined.

[0064] Specifically, since the Socket interface information that provides weight information to the automatic weight measurement program of scrap steel is sent at millisecond intervals, the judgment method to determine whether the weight of the scrap steel captured by the overhead crane and entering the steel bucket is stable is that the weight information provided by multiple Socket interface information between the Nth moment and the N-1th moment is changed relative to the previous weight information (i.e., the weight information at the Nth moment), and the change value gradually tends to be stable, until the weight information provided by the Socket interface information exceeding the preset threshold number and the weight information at the Nth moment are the same, then the difference is the weight information change amount after the weight information of the new scrap steel added to the steel bucket on the ground scale is stabilized, that is, the weight information provided by the Socket interface information of the preset threshold number is the weight information of the ground scale at the Nth moment. Preferably, the preset threshold value is 5, that is, the Socket interface information provides the weight information of the ground scale in real time, and in the process of loading the scrap steel into the steel bucket, the weight information provided by 5 consecutive Socket interface information is the same, which is regarded as stable, and the weight change is calculated, that is, the weight of the scrap steel newly added to the steel bucket can be obtained. Then query the scrap steel image of the most recent time that caused the change to identify the scrap steel type, obtain the scrap steel type and scrap steel weight newly added to the steel bucket, and finally automatically calculate the remaining weight requirement values ​​of each type of scrap steel.

[0065] Based on any of the above embodiments, in the method, determining the target scrap steel image corresponding to the change amount specifically includes:

[0066] Based on the Nth moment and the preset scrap steel transfer time, determine the scrap steel image collection moment;

[0067] The scrap steel image corresponding to the scrap steel collection moment is selected from the continuously collected scrap steel images as the target scrap steel image.

[0068] Specifically, after determining the change in weight information between the Nth moment and the N-1th moment, the moment of adding new scrap steel that causes this change is then determined. The moment is determined by rewinding the preset scrap steel transfer time based on the Nth moment. The preset scrap steel transfer time can be an estimated transfer time (the transfer time refers to the time between the start time when the scrap steel enters the corresponding range of the floor scale and the time when the scrap steel is placed in the steel bucket). The acquisition time of the target scrap steel image is determined as the Nth moment minus the estimated transfer time. For example, the Nth moment is the 458th second after the start of the operation, and the estimated transfer time is 17 seconds. Then, the transferred scrap steel that causes the jump in weight information at the Nth moment (i.e., the 458th second) should be captured in the target scrap steel image collected at 458-17=441 seconds. Therefore, the scrap steel image collected at the 441st second is selected from the continuously collected scrap steel images as the target scrap steel image corresponding to the change. Since the camera device will add a timestamp to each scrap steel image collected while continuously collecting scrap steel images to indicate the collection time of the scrap steel image, selecting the scrap steel image collected at the 441st second is equivalent to selecting the scrap steel image with a timestamp corresponding to the 441st second.

[0069] Based on any of the above embodiments, in this method, the preset scrap steel transfer time ranges from 0 seconds to 15 seconds.

[0070] Specifically, the preset scrap steel transfer time is further limited to a range of 0 seconds to 15 seconds.

[0071] Based on any of the above embodiments, in the method, before constructing the training sample set, the method further includes:

[0072] Using a camera device to collect images of different types of scrap steel in different environments, an initial scrap steel image set is obtained;

[0073] Each image in the initial scrap steel image set is rotated, scaled and flipped to obtain a plurality of sample scrap steel images.

[0074] Specifically, the collection of sample scrap steel images is explained here: the collection process can be to collect scrap steel images of different angles, different lighting conditions, different sizes and different types, so as to improve the generalization ability of the trained scrap steel type recognition model, ensure that the number of scrap steel samples of each category is roughly balanced, and avoid the trained scrap steel type recognition model being biased towards outputting scrap steel categories with a large number. The diversity of data is increased by means of rotation, scaling, flipping, etc., the robustness of the model to deformation, lighting changes, etc. is improved, and the image is standardized to speed up the convergence speed of the scrap steel type recognition model and improve the training stability.

[0075] Based on any of the above embodiments, in the method, the structure of the initial network to be trained is a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a third convolutional layer, a third pooling layer, a fourth convolutional layer and a fully connected layer connected in sequence, wherein the first convolutional layer includes 16 convolutional layers, the second convolutional layer includes 8 convolutional layers, the third convolutional layer includes 4 convolutional layers, and the fourth convolutional layer includes 2 convolutional layers.

[0076] Specifically, Figure 2 A schematic diagram of the structure of the initial network to be trained provided by an embodiment of the present invention, such as Figure 2 As shown, the structure of the initial network to be trained includes the first convolutional layer, the first pooling layer, the second convolutional layer, the second pooling layer, the third convolutional layer, the third pooling layer, the fourth convolutional layer and the fully connected layer connected in sequence, wherein the first convolutional layer includes 16 convolutional layers, the second convolutional layer includes 8 convolutional layers, the third convolutional layer includes 4 convolutional layers, and the fourth convolutional layer includes 2 convolutional layers. When training the initial network to be trained, the input layer of the network adopts an image size of 496×496×3, wherein, during the training process, first there are 16 consecutive convolutional layers with a stride of 1, 1 maximum pooling layer, then there are 8 consecutive convolutional layers with a stride of 1, 1 maximum pooling layer, then there are 4 consecutive convolutional layers with a stride of 1, 1 maximum pooling layer, and finally there are 2 consecutive convolutional layers with a stride of 1, and 3 fully connected layers.

[0077] In view of the lack of detailed scrap steel type and weight information in the scrap steel loading process in the scrap steel feeding link of the converter steelmaking production process, the weighing result relies heavily on manual experience, resulting in large errors, which seriously affects the subsequent steelmaking operation. The present invention provides a scrap steel weight automatic metering system based on CNN image recognition technology and weighing signals, which consists of three parts: a scrap steel plan management system, a scrap steel recognition subsystem and a scrap steel operation receiving system. The scrap steel plan management system is responsible for the acquisition of weighing signals, the acquisition of smelting plans, the generation of scrap steel plans, the loading and sending of the main business of the converter as the core. The system includes a task management module, a ratio setting module, a scrap steel metering module, and some basic floor scale maintenance, steel bucket maintenance, scrap steel ratio maintenance, report and other modules. The scrap steel recognition subsystem is responsible for identifying the scrap steel type information in the operation process according to the algorithm model generated by the scrap steel image training, and sending the scrap steel type information to the scrap steel plan management system. The scrap steel operation receiving subsystem is deployed on an overhead crane or a grab machine, and is responsible for receiving the operation plan sent by the scrap steel plan management system, and the loading operator performs the scrap steel loading operation according to the plan.

[0078] The smelting plan module of the scrap steel planning management system is responsible for interacting with the enterprise PES system to obtain the smelting plan operation information dynamically adjusted by the steelmaking department, which is presented in the form of a Gantt chart on the interface. The scrap steel plan can be generated with one click based on the smelting plan information.

[0079] The scrap steel plan management module in the scrap steel plan management system is responsible for adjusting the scrap steel type and weight ratio information of the plan according to the smelting steel type and sending the supply to the local scrap steel room. The status information of the plan can also be viewed through this module.

[0080] The scrap steel metering module of the scrap steel planning management system is responsible for the main business of scrap steel loading based on the real-time scrap steel image recognition results and weighing signals, including taking over the scale, setting the bucket number, completing the loading and sending to the converter. After the scale takes over the plan, it automatically obtains the image recognition results of the most recent time that affects this change by monitoring the changes in the scale indication, and automatically measures the scrap steel based on the recognition results.

[0081] The basic information maintenance module of the scrap steel planning management system is responsible for the addition, deletion, modification, query and maintenance of basic information such as overhead cranes, ground scales and steel buckets.

[0082] The scrap steel identification subsystem is developed based on the CNN scrap steel identification model. By fixing the high-definition camera at a suitable position near the workshop scale, the scrap steel identification subsystem identifies the collected scrap steel operation images during the operation, and sends the identification results and timestamp information to the scrap steel planning management system.

[0083] The scrap steel operation receiving subsystem is only responsible for reading the real-time information of the scrap steel plan taken over by the ground scale, and the scrap steel loading operator can check the remaining weight of the scrap steel loading in real time.

[0084] The beneficial effects of the embodiments of the present invention are as follows: the present invention consists of three parts: a scrap steel planning management system, a scrap steel identification subsystem and a scrap steel operation receiving subsystem. The scrap steel planning management system plays the role of the core brain, the scrap steel identification subsystem plays a key technical role, and the scrap steel operation receiving subsystem is responsible for facilitating the loading operator to understand the loading situation. The three parts work together to form an automatic scrap steel weight measurement system based on CNN image recognition technology and weighing signals. Through CNN image recognition technology, the system can more accurately identify different types of scrap steel, and combined with weighing signals, more accurate weight measurement can be achieved. The invention provides more accurate scrap steel weight data support for the steelmaking process, which is conducive to improving the quality of steel products. In addition, operators do not need to manually enter weight information, which reduces work intensity and improves work efficiency.

[0085] Based on any of the above embodiments, the embodiment of the present invention provides an automatic scrap steel weight measurement system based on CNN image recognition technology and weighing signals. Figure 3The schematic diagram of the structure of the automatic scrap steel weight measurement system based on CNN image recognition technology and weighing signal provided by the embodiment of the present invention is as follows: Figure 3 As shown in the figure, the scrap steel weight automatic measurement system based on CNN image recognition technology and weighing signal is composed of three systems responsible for different businesses that cooperate with each other through data interaction to complete the actual business. Among them, the scrap steel plan management system is the core, which is responsible for generating scrap steel plans according to the enterprise's smelting production plan, setting detailed scrap steel ratio information, and then starting loading operations. The loading operation link obtains the latest image recognition information that affects the weight change in real time by monitoring the changes in the ground scale signal to determine the scrap steel type information, so as to achieve the purpose of automatic calculation. After the loading is completed, it is sent to the converter to complete the main business process. The system also needs to be equipped with basic information maintenance modules such as steel buckets, as well as log recording and trend analysis modules. The scrap steel identification subsystem based on the CNN algorithm model is responsible for identifying the scrap steel type information during the operation process and sending the scrap steel type information to the scrap steel plan management system. The scrap steel operation receiving subsystem is deployed on the overhead crane or grab machine, which is responsible for receiving the operation plan sent by the scrap steel plan management system, and the loading operator performs scrap steel loading operations according to the plan.

[0086] Among them, when the CNN model is constructed, the input layer uses an image size of 496*496*3, 16 consecutive convolutional layers with a stride of 1, 1 maximum pooling layer, 8 consecutive convolutional layers with a stride of 1, 1 maximum pooling layer, 4 consecutive convolutional layers with a stride of 1, 1 maximum pooling layer, 2 consecutive convolutional layers with a stride of 1, and 3 fully connected layers.

[0087] When using scrap steel photo datasets to train CNN algorithm models, scrap steel images of different angles, different lighting conditions, different sizes and different types should be collected to improve the generalization ability of the model, ensure that the number of scrap steel samples of each category is roughly balanced, avoid the model being biased towards categories with a large number, increase the diversity of data through rotation, scaling, flipping and other means, improve the robustness of the model to deformation, lighting changes, etc., standardize the images, speed up the convergence of the model, and improve the stability of training. To prevent overfitting, Dropout, regularization and other technologies can be used to prevent the model from overfitting on the training set, and the cross-validation method can be used to evaluate the generalization ability of the model.

[0088] The HD camera should be installed at a suitable location near the weighbridge to ensure that it can capture images of scrap steel during loading. The clearer the captured images, the better the recognition effect. The scrap steel recognition subsystem recognizes the real-time photos of the scrap steel loading process captured by the HD camera based on the trained CNN algorithm model, and sends the prediction result information to the scrap steel planning management system.

[0089] The scrap steel planning and management system collects the weight information of the floor scale through the serial port. In the scrap steel metering link, the weight of the floor scale can be monitored in real time through the socket. During the loading process, if there are 5 or more consecutive socket weighing information, the reading is considered stable, and the weight change is calculated. Then, the scrap steel image recognition record of the most recent time that caused the change is queried, and automatic calculation is performed based on the identified scrap steel type information. In order to prevent the recognition subsystem from malfunctioning, a time threshold limit should be set, for example, only querying the scrap steel recognition records within the last 15 seconds. The system should be equipped with a basic information maintenance module, a historical record module and a curve analysis module.

[0090] In order to facilitate the scrap steel loading operators to view the detailed information of the scrap steel plan taken over on the ground scale in real time, the scrap steel operation receiving subsystem is deployed on the overhead crane or grab machine, which is responsible for receiving the operation plan sent by the scrap steel plan management system. The loading operators carry out scrap steel loading operations according to the plan.

[0091] The following describes an embodiment of the device of the present application, which can be used to execute the method for automatically measuring the weight of scrap steel in the above embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method for automatically measuring the weight of scrap steel in the above embodiment of the present application.

[0092] Based on any of the above embodiments, Figure 4 The present invention provides a schematic diagram of the structure of the automatic weight metering device for scrap steel provided in an embodiment of the present invention. The device is applied to a scrap steel recycling system, the scrap steel recycling system comprising the automatic weight metering device for scrap steel, an overhead crane, a floor scale and a steel bucket. The overhead crane can only grab one type of scrap steel at any one time during operation by an operator and place it in the steel bucket. The steel bucket is located on the floor scale. The floor scale is used to measure the total weight of the steel bucket and the scrap steel in the steel bucket. The automatic weight metering device for scrap steel is electrically connected to the floor scale, such as Figure 4 As shown, the device includes a collection unit 410, a determination unit 420, an identification unit 430, a measurement unit 440 and a display unit 450, wherein:

[0093] A collection unit 410 is specifically used to continuously collect the weight information of the floor scale and the image of the scrap steel transferred to the steel bucket;

[0094] A determination unit 420, configured to determine a change amount of the weight information and a target scrap steel image corresponding to the change amount if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment;

[0095] The recognition unit 430 is used to input the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained;

[0096] The measuring unit 440 is used to determine the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type identification result;

[0097] The display unit 450 is used to output display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

[0098] The automatic scrap steel weight measuring device provided by the embodiment of the present invention continuously collects the weight information of the floor scale and the scrap steel image transferred to the steel bucket. When the weight information of the floor scale changes, the change in weight information, i.e., the weight of the scrap steel transferred to the steel bucket at the Nth moment, is determined, and the target scrap steel image corresponding to the change in weight, i.e., the image of the scrap steel corresponding to the weight information jump at the Nth moment is determined. The target scrap steel image is input into the pre-trained scrap steel type recognition model to obtain the corresponding scrap steel type recognition result, so that the scrap steel type transferred to the steel bucket at the Nth moment and the weight of the scrap steel type transferred to the steel bucket can be obtained. Based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the weight demand values ​​of various types of scrap steel at the Nth moment can be updated, and the updated weight demand values ​​of various types of scrap steel are output and displayed for the operator to view and learn the remaining required weights of various types of scrap steel, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of multiple sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained. The automatic scrap steel weight measuring device provided by the present invention realizes the automatic scrap steel weight measuring function, can reduce the work intensity of on-site operators, and improve the scrap steel weight measuring accuracy.

[0099] Based on any of the above embodiments, in the device, the continuously collecting the weight information of the floor scale specifically includes:

[0100] The weight information of the scale is continuously collected through the serial interface and sent to the corresponding Socket interface.

[0101] Based on any of the above embodiments, in the device, the determining unit specifically includes:

[0102] If the number of socket interface messages received between the N-1th moment and the Nth moment that are different from the weight information at the N-1th moment and have the same weight information difference exceeds the preset threshold;

[0103] The difference is determined as a change in weight information between the Nth moment and the N-1th moment, and a target scrap steel image corresponding to the change is determined.

[0104] Based on any of the above embodiments, in the device, determining the target scrap steel image corresponding to the change amount specifically includes:

[0105] Based on the Nth moment and the preset scrap steel transfer time, determine the scrap steel image collection moment;

[0106] The scrap steel image corresponding to the scrap steel collection moment is selected from the continuously collected scrap steel images as the target scrap steel image.

[0107] Based on any of the above embodiments, in this method, the preset scrap steel transfer time ranges from 0 seconds to 15 seconds.

[0108] Based on any of the above embodiments, in the method, the collection of the plurality of sample scrap steel images specifically includes:

[0109] Using a camera device to collect images of different types of scrap steel in different environments, an initial scrap steel image set is obtained;

[0110] Each image in the initial scrap steel image set is rotated, scaled and flipped to obtain a plurality of sample scrap steel images.

[0111] Based on any of the above embodiments, in the method, the structure of the initial network to be trained is a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a third convolutional layer, a third pooling layer, a fourth convolutional layer and a fully connected layer connected in sequence, wherein the first convolutional layer includes 16 convolutional layers, the second convolutional layer includes 8 convolutional layers, the third convolutional layer includes 4 convolutional layers, and the fourth convolutional layer includes 2 convolutional layers.

[0112] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5As shown, the electronic device may include: a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logic instructions in the memory 530 to execute the automatic weight measurement method of scrap steel, which includes: continuously collecting the weight information of the floor scale and the scrap steel image transferred to the steel bucket; if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the target scrap steel image corresponding to the change amount; inputting the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of multiple sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained; determining the weight requirement values ​​of various types of scrap steel at the Nth moment based on the weight requirement values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type recognition result; and outputting display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

[0113] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0114] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the automatic scrap steel weight measurement method provided by the above methods, which method includes: continuously collecting the weight information of the floor scale and the scrap steel image transferred to the steel bucket; if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the target scrap steel image corresponding to the change amount; inputting the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of multiple sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained; determining the weight requirement values ​​of various types of scrap steel at the Nth moment based on the weight requirement values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type recognition result; and outputting display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for viewing by the operator.

[0115] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented by a processor to execute the automatic scrap steel weight measurement method provided by the above-mentioned methods when the computer program is executed, the method comprising: continuously collecting the weight information of the floor scale and the scrap steel image transferred to the steel bucket; if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the target scrap steel image corresponding to the change amount; inputting the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial network to be trained; determining the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type recognition result; and outputting display information indicating the weight demand values ​​of various types of scrap steel at the Nth moment for viewing by the operator.

[0116] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0117] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically measuring the weight of scrap steel, characterized in that: The invention is applied to a scrap steel recycling system, wherein the scrap steel recycling system comprises an automatic scrap steel weight measuring device, an overhead crane, a floor scale and a steel bucket. During the operation of the overhead crane by an operator, only one type of scrap steel can be grabbed and placed in the steel bucket at any one time. The steel bucket is located on the floor scale. The floor scale is used to measure the total weight of the steel bucket and the scrap steel in the steel bucket. The automatic scrap steel weight measuring device is electrically connected to the floor scale. The method comprises: Continuously collecting weight information of the floor scale and images of scrap steel transferred to the steel bucket; If there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the target scrap steel image corresponding to the change amount; Input the target scrap steel image into a pre-trained scrap steel type recognition model to obtain a scrap steel type recognition result corresponding to the target scrap steel image output by the scrap steel type recognition model, wherein the scrap steel type recognition model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial to-be-trained network for training; Determine the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type identification result; Outputting display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

2. The method for automatic weight measurement of scrap steel according to claim 1, characterized in that: The continuously collecting the weight information of the weighbridge specifically includes: The weight information of the scale is continuously collected through the serial interface and sent to the corresponding Socket interface.

3. The method for automatic weight measurement of scrap steel according to claim 2, characterized in that: If there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, determining the change amount of the weight information and the scrap steel image corresponding to the change amount specifically includes: If the number of socket interface messages received between the N-1th moment and the Nth moment that are different from the weight information at the N-1th moment and have the same weight information difference exceeds the preset threshold; The weight information difference is determined as a change in the weight information between the Nth moment and the N-1th moment, and a target scrap steel image corresponding to the change is determined.

4. The method for automatic weight measurement of scrap steel according to claim 1, characterized in that: Determining a target scrap steel image corresponding to the variation specifically includes: Based on the Nth moment and the preset scrap steel transfer time, determine the scrap steel image collection moment; The scrap steel image corresponding to the scrap steel collection moment is selected from the continuously collected scrap steel images as the target scrap steel image.

5. The method for automatically measuring the weight of scrap steel according to claim 4, characterized in that: The preset scrap steel transfer time ranges from 0 seconds to 15 seconds.

6. The method for automatic weight measurement of scrap steel according to claim 1, characterized in that: Before constructing the training sample set, the process further includes: Using a camera device to collect images of different types of scrap steel in different environments, an initial scrap steel image set is obtained; Each image in the initial scrap steel image set is rotated, scaled and flipped to obtain a plurality of sample scrap steel images.

7. The method for automatic weight measurement of scrap steel according to claim 6, characterized in that: The structure of the initial network to be trained is a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a third convolutional layer, a third pooling layer, a fourth convolutional layer and a fully connected layer connected in sequence, wherein the first convolutional layer includes 16 convolutional layers, the second convolutional layer includes 8 convolutional layers, the third convolutional layer includes 4 convolutional layers, and the fourth convolutional layer includes 2 convolutional layers.

8. An automatic weighing device for scrap steel, characterized in that: Applied to a scrap steel recycling system, the scrap steel recycling system comprises the scrap steel weight automatic metering device, an overhead crane, a floor scale and a steel bucket. The overhead crane can only grab one type of scrap steel and place it in the steel bucket at any one time during operation by an operator. The steel bucket is located on the floor scale. The floor scale is used to measure the total weight of the steel bucket and the scrap steel in the steel bucket. The scrap steel weight automatic metering device is electrically connected to the floor scale. The device comprises: A collection unit, specifically used to continuously collect the weight information of the floor scale and the image of the scrap steel transferred to the steel bucket; a determination unit, configured to determine, if there is a change between the weight information at the Nth moment and the weight information at the N-1th moment, a change amount of the weight information and a target scrap steel image corresponding to the change amount; an identification unit, configured to input the target scrap steel image into a pre-trained scrap steel type identification model, and obtain a scrap steel type identification result corresponding to the target scrap steel image output by the scrap steel type identification model, wherein the scrap steel type identification model is obtained by inputting a training sample set consisting of a plurality of sample scrap steel images and corresponding scrap steel type labels into an initial to-be-trained network for training; A measuring unit, used for determining the weight demand values ​​of various types of scrap steel at the Nth moment based on the weight demand values ​​of various types of scrap steel at the N-1th moment, the change amount and the scrap steel type identification result; The display unit is used to output display information indicating the weight requirement values ​​of various types of scrap steel at the Nth moment for the operator to view.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the operations performed by the method for automatic weight measurement of scrap steel as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: The electronic device includes one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the operations performed by the automatic scrap steel weight measurement method as described in any one of claims 1 to 7.

Images