A method, device and equipment for identifying information of a splashing slag protection furnace and a storage medium

Abstract

The present application provides a kind of splashing slag protects furnace and jumps slag information identification method, device, equipment and storage medium, the method is by obtaining initial image and to be identified image, based on the initial network model of labeled initial image is trained, obtains the output result of jump slag identification model, the output result includes multiple different labels, different label in output result is assigned to different pixel value, to obtain target jump slag image, identify target jump slag area in target jump slag image, to obtain the jump slag information of to be identified image;By image construction image recognition model, the intelligent identification of to be identified jump slag image is carried out, realizes the identification of converter jump slag particle, improves its identification accuracy, reduces the input of manpower, and provides reliable data source for other intelligent processing.

Description

Technical Field

[0001] This application relates to the field of iron and steel smelting technology, specifically to a method, device, equipment, and storage medium for identifying slag splashing and furnace jumping information. Background Technology

[0002] In the iron and steel smelting industry, furnace life is a comprehensive technical and economic indicator for converter steelmaking. The mechanical erosion and chemical corrosion of the furnace lining by high-temperature, highly oxidizing slag are the main causes of lining damage. Slag splashing protection technology is a commonly used method to improve furnace life. In this process, a high-speed nitrogen jet is injected through a top-blown oxygen lance to impact the remaining high-melting-point slag in the molten pool, causing the slag to splash and coat the entire surface of the converter lining, forming a protective slag layer. In actual production, the slag splashing situation needs to be monitored in real time during nitrogen blowing to adjust the lance height and nitrogen blowing speed.

[0003] Currently, in the steel smelting industry, the monitoring of slag jumping in slag splashing furnace protection mainly relies on manual methods, which have relatively low accuracy and efficiency. It is impossible to quickly and accurately determine the number, size, and location of jumping slag, and it cannot provide data support for subsequent intelligent processing. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the present invention provides a method, device, equipment and storage medium for identifying slag jumping information in slag splashing furnace protection, so as to solve the technical problems mentioned above, which are that the monitoring of slag jumping in the current slag splashing furnace protection relies on manual labor, has low accuracy and efficiency, cannot quickly and accurately determine the quantity, size and location of slag jumping, and cannot provide data support for subsequent intelligent processing.

[0005] This invention proposes a method for identifying slag splashing and furnace jumping information. The method includes: acquiring an initial image and an image to be identified; training an initial network model based on the labeled initial image to obtain a slag jumping identification model; inputting the image to be identified into the slag jumping identification model to obtain the output result of the slag jumping identification model, the output result including multiple different labels; assigning different pixel values ​​to the image regions where different labels are located in the output result to obtain a target slag jumping image; identifying the target slag jumping region in the target slag jumping image to obtain the slag jumping information of the image to be identified.

[0006] In one embodiment of this application, training an initial network model based on annotated initial images to obtain a "scumbag identification model" includes: obtaining scumbag information based on scumbag images and obtaining background information based on non-scumbag images, wherein the initial images include scumbag images and non-scumbag images; cropping the initial images to obtain sample images; generating a first label based on the scumbag information and a second label based on the background information; and annotating the sample images based on the first label and the second label; obtaining a training sample dataset based on the annotated sample images; and training the initial network model based on the training sample dataset to obtain the scumbag identification model.

[0007] In one embodiment of this application, cropping the initial image to obtain a sample image includes: acquiring preset cropping information and size information of the initial image, the size information including an initial width and an initial height; segmenting the initial image based on the preset cropping information, the initial width, and the initial height to obtain multiple cropped images, and determining the cropped images as sample images.

[0008] In one embodiment of this application, labeling the sample image based on the slag jumping information and the background information includes: identifying image information of the sample image, the image information including slag jumping information and background information; determining the area where the slag jumping information is located as a slag jumping block and the area where the background information is located as a background block; labeling the slag jumping block based on a preset first label and labeling the background block based on a preset second label.

[0009] In one embodiment of this application, obtaining a training sample dataset based on labeled sample images includes: performing data transformation on the labeled sample images to obtain virtual sample images, wherein the data transformation includes at least one of the following transformation methods: flip transformation, scaling transformation, translation transformation, contrast transformation, noise perturbation, and Gaussian transformation; and determining the labeled sample images and the virtual sample images as the training sample dataset.

[0010] In one embodiment of this application, different pixel values ​​are assigned to the image regions where different labels are located in the output results to obtain a target scumbag image, including: receiving the output image of the scumbag identification model, the output image including a preset first label and a preset second label; assigning a first pixel value to the region where the first label is located, and assigning a second pixel value to the region where the second label is located, to obtain a target scumbag image.

[0011] In one embodiment of this application, identifying a target debris-jumping region in the target debris-jumping image includes: determining any debris-jumping block in the target debris-jumping region as a target debris-jumping region, obtaining the individual area of ​​the target debris-jumping block, wherein the target debris-jumping region includes at least one debris-jumping block; comparing the individual area with a preset area threshold, and determining the target debris-jumping region as a valid debris-jumping block when the individual area is greater than a preset minimum area and less than a preset maximum area; traversing each debris-jumping block of the target debris-jumping region to obtain all valid debris-jumping blocks, and determining the region where all the valid debris-jumping blocks are located as the target debris-jumping region of the image to be identified.

[0012] In one embodiment of this application, after determining the area where all the legal slag-jumping blocks are located as the target slag-jumping area of ​​the image to be identified, the method further includes: calculating the sum of the areas of all the legal slag-jumping blocks based on the individual area of ​​each slag-jumping block, and determining the sum of the areas as the slag-jumping area of ​​the image to be identified; counting the number of the legal slag-jumping blocks to obtain the number of slag-jumping blocks in the image to be identified; recording the location information of each slag-jumping block to obtain the slag-jumping distribution of the image to be identified; and determining the slag-jumping area, the number of slag-jumping blocks, and the slag-jumping distribution information as the slag-jumping information of the image to be identified.

[0013] This application provides a slag splashing protection furnace slag jumping identification device, the device comprising: an information acquisition module for acquiring an initial image and an image to be identified; a model construction module for training an initial network model based on the labeled initial image to obtain a slag jumping identification model; an image recognition module for inputting the image to be identified into the slag jumping identification model to obtain the output result of the slag jumping identification model, the output result including multiple different labels; an image parsing module for assigning different pixel values ​​to the image regions where different labels are located in the output result to obtain a target slag jumping image; and a slag jumping information determination module for identifying the target slag jumping region in the target slag jumping image to obtain the slag jumping information of the image to be identified.

[0014] In one embodiment of this application, the model building module includes: an image information recognition submodule, used to obtain jumping information based on jumping images and background information based on non-jumping images, wherein the initial image includes jumping images and non-jumping images; an image cropping submodule, used to crop the initial image to obtain a sample image; an image annotation submodule, used to generate a first label based on the jumping information, generate a second label based on the background information, and annotate the sample image based on the first label and the second label; and a model training submodule, used to obtain a training sample dataset based on the annotated sample image, and train the initial network model based on the training sample dataset to obtain a jumping recognition model.

[0015] In one embodiment of this application, the "skipping debris information determination module" includes: an image receiving submodule, configured to receive the output image of the skimming debris recognition model, the output image including a preset first label and a preset second label; an image parsing submodule, configured to assign a first pixel value to the area where the first label is located and assign a second pixel value to the area where the second label is located, to obtain a target skimming debris image; a skimming debris region determination submodule, configured to calculate the area of ​​each skimming debris block to obtain a legal skimming debris block, and determine the target skimming debris region of the image to be identified based on the legal skimming debris blocks; and a skimming debris information acquisition submodule, configured to calculate the skimming debris area of ​​the target skimming debris region, count the number of legal skimming debris blocks, and record the skimming debris distribution of multiple skimming debris blocks.

[0016] This application proposes an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device enables the slag splashing protection furnace jumping information identification method as described above.

[0017] This application proposes a computer-readable storage medium, characterized in that it stores a computer program thereon, which, when executed by a computer processor, causes the computer to perform the slag splashing protection furnace jumping information identification method as described above.

[0018] The beneficial effects of this invention are as follows: This invention provides a method, apparatus, device, and storage medium for identifying slag splashing and furnace jumping information. The method acquires jumping images, non-jumping images, and images to be processed. Based on the non-jumping image, a background image of the initial image is obtained. The parts of the jumping image that are identical to the background image are identified as background information, and the parts of the jumping image that differ from the background image are identified as jumping information. The jumping images are then labeled based on the background information and the jumping information to obtain a training sample dataset. An initial network model is trained based on the training sample dataset to obtain a jumping recognition model. The image to be processed is input into the jumping recognition model to identify the jumping state of the image. By constructing an image recognition model from the images, intelligent recognition of the jumping images to be identified is achieved, realizing the identification of converter jumping slag particles, improving its recognition accuracy, reducing manpower input, and providing a reliable data source for other intelligent processing.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0021] Figure 1 This is a schematic diagram illustrating the implementation environment of the slag splashing protection furnace jumping information identification method, as shown in an exemplary embodiment of this application;

[0022] Figure 2 This is a flowchart illustrating an exemplary embodiment of the present application of a method for identifying slag splashing and furnace jumping information;

[0023] Figure 3 This is an example of the original image of slag jumping in the slag splashing protection furnace slag jumping information identification method shown in an exemplary embodiment of this application;

[0024] Figure 4 This is an example of an annotated image of the original image in the slag splashing protection furnace jumping information identification method shown in an exemplary embodiment of this application;

[0025] Figure 5 This is an example of the identification and segmentation effect in the slag splashing protection furnace jumping information identification method shown in an exemplary embodiment of this application;

[0026] Figure 6 This is a block diagram illustrating a slag splashing protection furnace jumping slag identification device, as shown in an exemplary embodiment of this application;

[0027] Figure 7 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0028] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0029] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0030] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0031] First, it's important to note that UNet is a variant of FCN, arguably the most commonly used and simplest segmentation model. It's simple, efficient, easy to understand, easy to build, and can be trained on small datasets. FCN (Fully Convolutional Networks for Semantic Segmentation), as the name suggests, replaces the original fully connected layers with convolutional layers.

[0032] The Transformer is an end-to-end NLP paradigm. This model abandons the traditional sequential structure of RNNs and employs a self-attention mechanism to enable parallel training and the acquisition of global information. Typically, Transformer models are trained on large text corpora and then fine-tuned on smaller, specific tasks to achieve better computational efficiency and accuracy. Vision Transformer can be seen as a graphical version of the Transformer, directly transferring the standard Transformer model to the image domain with minimal modifications.

[0033] Figure 1 This is a schematic diagram illustrating the implementation environment of the slag splashing protection furnace jumping information identification method, as shown in an exemplary embodiment of this application. Figure 1As shown, the implementation environment of the slag splashing protection furnace slag jumping information identification method includes an image acquisition device 101 and a computer device 102. The image acquisition device 101 is used to acquire original slag jumping images. This image acquisition device can be any device that can acquire image information, such as a camera or video camera; this application does not impose any restrictions on it. The computer device 102 is specifically configured with a corresponding network algorithm and preset training rules, used to train the slag jumping identification model based on the original slag jumping images acquired by the image acquisition device 101, to obtain a target slag jumping identification model, and used to identify the slag jumping state of the image to be identified. Furthermore, the computer device 102 can be at least one of a desktop graphics processing unit (GPU) computer, a GPU computing cluster, a neural network computer, etc., or it can be an intelligent processor integrated into the current vehicle; this application also does not impose any restrictions on this.

[0034] Figure 2 This is a flowchart illustrating an exemplary embodiment of the present application of a method for identifying slag splashing and furnace jumping information.

[0035] like Figure 2 As shown, in an exemplary embodiment, the slag splashing protection furnace jumping information identification method includes at least steps S210 to S240, which are described in detail below:

[0036] Step S210: Obtain the initial image and the image to be recognized.

[0037] Figure 3 This is an example of the original image of slag splashing in the slag splashing protection furnace slag jumping information identification method shown in an exemplary embodiment of this application; such as Figure 3 As shown, there are scattered white bright spots in the image, which are the slag particles, while the remaining black areas are the background area.

[0038] In one embodiment of this application, an infrared thermal imaging sensor is used as an image acquisition device. Since the temperature at the slag-jumping point is significantly higher than the temperature of the background area, points with a temperature higher than or equal to a preset temperature are identified as slag-jumping points and displayed in white, while the remaining areas with temperatures lower than the preset temperature are identified as background areas and displayed in black. Figure 3 The image shown is the original image of the slag, i.e., the initial image.

[0039] Step S220: Train the initial network model based on the labeled initial image to obtain the scumbag recognition model.

[0040] In one embodiment of this application, training an initial network model based on annotated initial images to obtain a "scumbag identification model" includes: obtaining scumbag information based on scumbag images and obtaining background information based on non-scumbag images, wherein the initial images include scumbag images and non-scumbag images; cropping the initial images to obtain sample images; generating a first label based on the scumbag information and a second label based on the background information; and annotating the sample images based on the first and second labels; obtaining a training sample dataset based on the annotated sample images; and training the initial network model based on the training sample dataset to obtain the scumbag identification model.

[0041] In one embodiment of this application, after obtaining the initial image, based on prior knowledge, it can be identified whether the initial image contains a slag-jumping area. If the initial image contains a slag-jumping area, the image information of the slag-jumping area in the initial image is determined as slag-jumping information; otherwise, the image information of the non-slag-jumping area and the image information of the initial image that does not contain slag-jumping information are determined as background information.

[0042] In one embodiment of this application, cropping an initial image to obtain a sample image includes: acquiring preset cropping information and size information of the initial image, the size information including initial width and initial height; segmenting the initial image based on the preset cropping information, initial width, and initial height to obtain multiple cropped images, and determining the cropped images as sample images.

[0043] In one specific embodiment of this application, taking an original image with a resolution of H×W and preset cropping information of a cropped image height of v and an image width of h as an example, based on the pixel information of the original image, it can be known that the original image height is H, that is, the initial image height is H, and the original image width is W, that is, the initial image width is W. Therefore, based on its height value H, it is cropped vertically into v parts, and based on its width value W, it is divided horizontally into h parts. Based on this, the original image can be cropped into v×h small images, and the cropped small images are stored as sample images in the computer memory. The formula for calculating the number of small images is as follows:

[0044] Equation (1),

[0045] Where X is the number of sub-images, H is the height of the original image, v is the height of the sub-image, W is the width of the original image, and h is the width of the sub-image.

[0046] It should be noted that by cropping the initial image into smaller sample images, the amount of computation required in the subsequent model training process can be greatly reduced, thereby effectively reducing the memory requirements when training the neural network.

[0047] In one embodiment of this application, the annotation of a sample image based on slag jumping information and background information includes: identifying image information of the sample image, the image information including slag jumping information and background information; determining the area where the slag jumping information is located as a slag jumping block and the area where the background information is located as a background block; annotating the slag jumping block based on a preset first label and annotating the background block based on a preset second label.

[0048] It should be noted that after obtaining the initial image, further identification and judgment are needed to facilitate subsequent image annotation. The main purpose is to distinguish between non-skimming and non-skimming regions in the image. Therefore, different labels can be assigned to images of skipping and non-skimming regions, based on differences in pixel values ​​to differentiate the image region's attribute as skipping or non-skimming. It should be noted that assigning different labels to images as described above is merely an illustrative method for distinguishing skipping regions and does not limit the specific implementation of this application.

[0049] In one embodiment of this application, (H, W, ) represents a labeled image, where H represents the image height (pixels) and W represents the image width (pixels). The annotation values ​​are different between the slag particles and the background. That is, the annotation values ​​of the slag area and the background area are different in the initial image. The initial image is modified to a dual-channel image, namely channel 0 and channel 1. If the pixel is assigned to the slag particle, then the value of channel 0 is 1 and the value of channel 1 is 0. If the pixel is assigned to the background, then the value of channel 0 is 0 and the value of channel 1 is 1. Therefore, the modified image can be represented as (H, W, 2).

[0050] In one specific embodiment of this application, a training sample dataset is obtained through the above steps. The pixel values ​​representing the background in the images of the training sample dataset are modified to (0, 1), and the pixel values ​​representing the slag particles in the images are modified to (1, 0). Then, the images labeled according to this standard are input into the model for training according to a preset number of training batches. The specific representation is as follows:

[0051] Equation (2),

[0052] Where pixelvalue is the displayed pixel value, (1,0) represents the scraggly area, and (0,1) represents the background area.

[0053] In one specific embodiment of this application, obtaining a training sample dataset based on labeled sample images includes: performing data transformation on the labeled sample images to obtain virtual sample images, wherein the data transformation includes at least one of the following transformation methods: flip transformation, scaling transformation, translation transformation, contrast transformation, noise perturbation, and Gaussian transformation; and determining the labeled sample images and virtual sample images as the training sample dataset.

[0054] In one specific embodiment of this application, the labeled sample images are transformed to obtain a training sample dataset, i.e., data augmentation is performed on the labeled dataset. The transformation methods include, but are not limited to: affine transformation, elastic transformation, horizontal flipping, vertical flipping, sharpening, perspective transformation, piecewise affine transformation, and random pixel discarding. Any one of the above processing methods can be selected to transform the training images, or multiple data processing methods can be randomly selected and randomly combined to process the training images.

[0055] In one specific embodiment of this application, the labeled sample images are transformed using two data transformation methods: horizontal flipping and vertical flipping, to obtain a training sample dataset. First, the labeled sample images are horizontally flipped to obtain a first dataset; then, all data in the first dataset are vertically flipped to obtain a second dataset; finally, the labeled sample images, the images from the first dataset, and the images from the second dataset are combined to construct the sample dataset.

[0056] It should be noted that there are many ways to obtain new data related to the sample data by transforming the sample data. The various data transformation methods proposed above are only illustrative examples. In actual data processing, one or more of the above data transformation methods can be used to process the sample data, or multiple data processing methods can be organically combined to transform the sample data. This application does not impose any restrictions on the methods of data transformation during data processing. In addition, processing the sample data based on the data transformation methods proposed above can achieve the purpose of data augmentation, effectively increasing the total number of training samples in the training dataset.

[0057] Furthermore, it should be noted that the initial network model used to train the model for identifying slackers includes, but is not limited to, neural network models for image segmentation such as the UNet series and Vision Transformer (ViT). Other image segmentation networks can be used instead. This application does not impose any restrictions on the type of initial network model used to train the model for identifying slackers.

[0058] Figure 4 This is an example of an annotated image of the original image in the slag splashing protection furnace jumping information identification method shown in an exemplary embodiment of this application.

[0059] In one embodiment of this application, when labeling the images used to train the initial network model, images with varying degrees of jumping and images without jumping are selected. Pixels belonging to jumping and background in the training images are labeled with two different tags. The labeled images are then stored in a computer memory. The labeled images are then... Figure 4 As shown.

[0060] Step S230: Input the image to be identified into the scumbag identification model to obtain the output result of the scumbag identification model, which includes multiple different labels.

[0061] In one embodiment of this application, the output of the scumbag identification model includes a first label and a second label, wherein the area where the second label is located is a large, continuous area, while the image area where the second label is located consists of scattered blocks interspersed within the first label.

[0062] Step S240: Assign different pixel values ​​to the image regions where different labels are located in the output results to obtain the target slag image.

[0063] In one embodiment of this application, different pixel values ​​are assigned to the image regions where different labels are located in the output results to obtain a target scumbag image, including: receiving the output image of the scumbag recognition model, the output image including a preset first label and a preset second label; assigning a first pixel value to the region where the first label is located, and assigning a second pixel value to the region where the second label is located to obtain a target scumbag image.

[0064] In one specific embodiment of this application, taking the assignment of a value of 255 to the image of the "jumping debris" region (i.e., the preset first label region) and a value of 0 to the image of the background region (i.e., the preset second label region) as an example, the output image, i.e., the output data, of the jumping debris recognition model is first obtained. Then, the output data is decoded to obtain a grayscale image, which is represented as follows:

[0065] Equation (3),

[0066] Where pixelvalue is the displayed pixel value, 255 is the value assigned to the area of ​​slag particles in the image, and 0 is the value assigned to the background area in the image.

[0067] In one embodiment of this application, identifying a target debris-jumping region in a target debris-jumping image includes: determining any debris-jumping block in the target debris-jumping region as a target debris-jumping region, obtaining the individual area of ​​the target debris-jumping block, wherein the target debris-jumping region includes at least one debris-jumping block; comparing the individual area with a preset area threshold, wherein when the individual area is greater than a preset minimum area and less than a preset maximum area, the target debris-jumping region is determined as a legal debris-jumping block; traversing each debris-jumping block of the target debris-jumping region to obtain all legal debris-jumping blocks, and determining the region where all legal debris-jumping blocks are located as the target debris-jumping region of the image to be identified.

[0068] It should be noted that, due to the randomness of slag jumping, the slag jumping area in the image is usually formed by multiple unconnected slag jumping blocks. Therefore, when calculating the area of ​​the slag jumping area, it is necessary to determine the area of ​​each slag jumping block separately, and obtain the total area based on the areas of each small slag jumping block, which is the area of ​​the slag jumping area in the image. Furthermore, since the area of ​​the slag jumping blocks is uncertain, for ease of calculation and to better reflect the needs of actual production and daily life, slag jumping blocks that are too large or too small can be removed, resulting in slag jumping blocks of suitable size as the components of the slag jumping area in the image.

[0069] In one embodiment of this application, regions identified as "slag jumpers" in the image are filtered to remove areas that are too large or too small, preventing potential misidentification from interfering with subsequent slag jumper counts. Two thresholds, "highbond" (too large area) and "lowbond" (too small area), are set to filter the initial slag jumper blocks. It should be noted that these thresholds should be modified according to specific production conditions; this application does not impose any restrictions on the specific thresholds. The filtering steps are as follows:

[0070] First, find the outline of all regions in the image with a pixel value of 255, that is, the outline of the slag block;

[0071] Then, calculate the area within the outline of each of the identified slag-jumping blocks in the image, that is, determine the individual area of ​​each slag-jumping block;

[0072] Next, the obtained unit area is compared with the preset excessively large area and excessively small area, and the contour data with an area smaller than lowbond and larger than highbond are deleted.

[0073] Finally, after screening the individual areas of each slag-jumping block, the slag-jumping blocks that meet the requirements are determined as target slag-jumping blocks. The sum of the areas of each target slag-jumping block is calculated to obtain the slag-jumping area of ​​the image to be identified, and the relevant data of each target slag-jumping block are statistically analyzed.

[0074] Step S250: Identify the target debris-jumping region in the target debris-jumping image to obtain debris-jumping information of the image to be identified.

[0075] In one embodiment of this application, after determining the area where all legal slag-jumping blocks are located as the target slag-jumping area of ​​the image to be identified, the method further includes: calculating the sum of the areas of all legal slag-jumping blocks based on the individual area of ​​each slag-jumping block, and determining the sum of the areas as the slag-jumping area of ​​the image to be identified; counting the number of legal slag-jumping blocks to obtain the number of slag-jumping blocks in the image to be identified; recording the location information of each slag-jumping block to obtain the slag-jumping distribution of the image to be identified; and determining the slag-jumping area, the number of slag-jumping blocks, and the slag-jumping distribution information as the slag-jumping information of the image to be identified.

[0076] In one specific embodiment of this application, the relevant data of each target debris-jumping block are statistically analyzed to obtain the area size, number of target debris-jumping areas, and location information of each target debris-jumping area in the image to be identified, and a debris-jumping record table is generated based on the above relevant information.

[0077] Furthermore, it should be noted that the above embodiments are based on the slag jumping information obtained after screening the individual areas of each slag jumping block. In actual production and daily life, an initial slag jumping information record table identical to the above-mentioned slag jumping record table can also be obtained based on the information of each unscreened slag jumping block. Therefore, a slag jumping record table can be established based on the screened legal slag jumping blocks according to actual needs to obtain the slag jumping information of the image to be identified; or an initial slag jumping record table can be established based on the information of all unscreened slag jumping blocks. This application does not impose any restrictions on whether or not slag jumping blocks are screened in the slag jumping information table, or on the categories of slag jumping blocks in the slag jumping information table.

[0078] In one embodiment of this application, after generating the slag jumping record table, the method further includes: sending the slag jumping record table to the output terminal to display the slag jumping status of the image to be identified.

[0079] In one embodiment of this application, a real-time slag-jumping image of a converter in progress is used as the image to be identified. After obtaining the slag-jumping status and slag-jumping record table of the image to be identified through the above steps, the data of the slag-jumping record table is sent to the terminal. After receiving the data of the slag-jumping record table, the terminal generates corresponding charts based on the relevant data and outputs the relevant charts in a visual form through the terminal's dashboard, so that relevant personnel can more intuitively determine the slag-jumping situation of the converter and take corresponding measures in a timely manner based on its specific slag-jumping situation.

[0080] In one specific embodiment of this application, the mobile terminal can generate a real-time debris-jumping map based on the location information of each target debris-jumping block in the debris-jumping record table data, so as to represent the distribution status of each debris-jumping block.

[0081] It should be noted that the aforementioned terminal is a device with a visual dashboard, which can be a smart dashboard, a tablet computer, or any other device that can receive data and generate and display charts based on the received data. This application does not impose any restrictions on the terminal.

[0082] Figure 5 This is an example of the identification and segmentation effect in the slag splashing protection furnace jumping information identification method shown in an exemplary embodiment of this application. For example... Figure 5 As shown, the image contains bright color blocks and large black areas, where the bright color blocks are the splash areas and the black areas are the background areas.

[0083] In one embodiment of this application, a debris-jumping recognition model is first constructed based on a large amount of sample data. Then, the target image to be recognized is input into the debris-jumping recognition model to obtain a recognition result. Next, the recognition result is decoded, i.e., different pixel values ​​are assigned to debris-jumping and non-debris-jumping regions respectively, resulting in a debris-jumping image with bright spots. Finally, the area of ​​each debris-jumping region in the image is calculated, and debris-jumping blocks with areas exceeding a preset range are removed, resulting in... Figure 5 The final image of the slag heap shown.

[0084] Figure 6 This is a block diagram illustrating a slag splashing protection and slag jumping identification device, as shown in an exemplary embodiment of this application. The device can be applied to... Figure 1 The implementation environment shown is not limited to this embodiment. This device can also be applied to other exemplary implementation environments and specifically configured in other devices.

[0085] like Figure 6 As shown, the exemplary slag splashing protection furnace jumping slag identification device includes: an information acquisition module 610, a model construction module 620, an image recognition module 630, an image analysis module 640, and a jumping slag information determination module 650.

[0086] The system includes: an information acquisition module 610 for acquiring an initial image and an image to be identified; a model construction module 620 for training an initial network model based on the labeled initial image to obtain a scumbag identification model; an image recognition module 630 for inputting the image to be identified into the scumbag identification model to obtain the output result of the scumbag identification model, which includes multiple different labels; an image parsing module 640 for assigning different pixel values ​​to the image regions containing different labels in the output result to obtain the target scumbag image; and a scumbag information determination module 650 for identifying the target scumbag region in the target scumbag image to obtain the scumbag information of the image to be identified.

[0087] In addition, the model building module includes: an image information recognition submodule, used to obtain jumping information based on jumping images and background information based on non-jumping images, the initial images including jumping images and non-jumping images; an image cropping submodule, used to crop the initial images to obtain sample images; an image annotation submodule, used to generate a first label based on jumping information and a second label based on background information, and to annotate the sample images based on the first and second labels; and a model training submodule, used to obtain a training sample dataset based on the annotated sample images, and to train the initial network model based on the training sample dataset to obtain a jumping recognition model. The "jumping debris information determination module" includes: an image receiving submodule, used to receive the output image of the jumping debris recognition model, the output image including a preset first label and a preset second label; an image parsing submodule, used to assign a first pixel value to the area where the first label is located and assign a second pixel value to the area where the second label is located, so as to obtain the target jumping debris image; a jumping debris region determination submodule, used to calculate the area of ​​each jumping debris block to obtain a legal jumping debris block, and determine the target jumping debris region of the image to be recognized based on the legal jumping debris blocks; and a jumping debris information acquisition submodule, used to calculate the jumping debris area of ​​the target jumping debris region, count the number of legal jumping debris blocks, and record the jumping debris distribution of multiple jumping debris blocks.

[0088] It should be noted that the slag splashing and slag jumping identification device provided in the above embodiments and the slag splashing and slag jumping information identification method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the slag splashing and slag jumping identification device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0089] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, enable the electronic device to implement the slag splashing and furnace protection slag jumping information identification method provided in the above embodiments.

[0090] Figure 7 A schematic diagram of a computer system suitable for implementing the embodiments of this application is shown. It should be noted that... Figure 7 The computer system 700 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0091] like Figure 7 As shown, the computer system 700 includes a Central Processing Unit (CPU) 701, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 702 or programs loaded from storage portion 708 into Random Access Memory (RAM) 703, such as performing the methods described in the above embodiments. The RAM 703 also stores various programs and data required for system operation. The CPU 701, ROM 702, and RAM 703 are interconnected via a bus 704. An Input / Output (I / O) interface 705 is also connected to the bus 704.

[0092] The following components are connected to I / O interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to I / O interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 710 as needed so that computer programs read from it can be installed into storage section 708 as needed.

[0093] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by central processing unit (CPU) 701, it performs various functions defined in the system of this application.

[0094] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0095] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0096] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0097] Another aspect of this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the slag splashing and furnace protection slag jumping information identification method as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not be assembled into the electronic device.

[0098] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the slag splashing and furnace jumping information identification method provided in the various embodiments above.

[0099] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for identifying slag splashing and furnace jumping information, characterized in that, The method includes: Obtain the initial image and the image to be recognized; The initial network model is trained based on the labeled initial images to obtain the scumbag recognition model; The image to be identified is input into the scumbag identification model to obtain the output result of the scumbag identification model, and the output result includes multiple different labels; Different pixel values ​​are assigned to the image regions containing different labels in the output results to obtain the target slag image; Identify the target debris-jumping region in the target debris-jumping image to obtain debris-jumping information of the image to be identified. The debris-jumping information includes debris-jumping area, number of debris-jumping blocks, and debris-jumping distribution information. Statistical analysis is performed on the relevant data of each target debris-jumping area to obtain the area size, number of target debris-jumping areas, and location information of each target debris-jumping area in the image to be identified, and a debris-jumping record table is generated. The target debris-jumping area includes at least one debris-jumping block, and the target debris-jumping block is each debris-jumping block that meets the requirements after filtering the individual area of ​​each debris-jumping block. Send the debris-jumping record table to the output terminal to display the debris-jumping status of the image to be identified; Based on the location information of each target debris-jumping block in the debris-jumping record table data, the mobile terminal generates a real-time debris-jumping map to represent the distribution status of each debris-jumping block.

2. The method for identifying slag splashing and furnace jumping information according to claim 1, characterized in that, The initial network model is trained based on the labeled initial images to obtain the "jumping scum" identification model, including: The initial image includes images of people jumping over slag and images of people not jumping over slag. The initial image includes images of people jumping over slag and images of people not jumping over slag. The initial image is cropped to obtain a sample image. A first label is generated based on the jump slag information, a second label is generated based on the background information, and the sample image is labeled based on the first label and the second label. A training sample dataset is obtained based on the labeled sample images, and the initial network model is trained based on the training sample dataset to obtain the scumbag identification model.

3. The method for identifying slag splashing and furnace jumping information according to claim 2, characterized in that, Cropping the initial image to obtain a sample image includes: Obtain preset cropping information and the size information of the initial image, wherein the size information includes the initial width and initial height; The initial image is segmented based on the preset cropping information, the initial width, and the initial height to obtain multiple cropped images, and the cropped images are determined as sample images.

4. The method for identifying slag splashing and furnace jumping information according to claim 2, characterized in that, The sample image is labeled based on the slag jumping information and the background information, including: Identify the image information of the sample image, the image information including slag information and background information; The area where the slag jumping information is located is defined as the slag jumping block, and the area where the background information is located is defined as the background block; The slag-jumping area is labeled based on a preset first label, and the background area is labeled based on a preset second label.

5. The method for identifying slag splashing and furnace jumping information according to claim 2, characterized in that, The training sample dataset is obtained based on the labeled sample images, including: The labeled sample image is subjected to data transformation to obtain a virtual sample image. The data transformation includes at least one of the following transformation methods: flip transformation, scaling transformation, translation transformation, contrast transformation, noise perturbation, and Gaussian transformation. The labeled sample images and the virtual sample images are determined as the training sample dataset.

6. The method for identifying slag splashing and furnace jumping information according to any one of claims 1-5, characterized in that, Assigning different pixel values ​​to the image regions containing different labels in the output results to obtain the target slag image, including: Receive the output image of the scumbag identification model, the output image including a preset first label and a preset second label; The region containing the first label is assigned a first pixel value, and the region containing the second label is assigned a second pixel value to obtain the target image of the slag.

7. The method for identifying slag splashing and furnace jumping information according to claim 6, characterized in that, Identifying the target debris-jumping region in the target debris-jumping image includes: Any slag-jumping block in the target slag-jumping area is determined as the target slag-jumping area, and the individual area of ​​the target slag-jumping block is obtained. The target slag-jumping area includes at least one slag-jumping block. The area of ​​the single unit is compared with a preset area threshold. When the area of ​​the single unit is greater than a preset minimum area and less than a preset maximum area, the target slag jumping area is determined as a legal slag jumping block. Traverse each block of the target debris-jumping region to obtain all legal debris-jumping blocks, and determine the region where all the legal debris-jumping blocks are located as the target debris-jumping region of the image to be identified.

8. The method for identifying slag splashing and furnace jumping information according to claim 7, characterized in that, The methods for determining the slag spill area include: The sum of the areas of all legal slag-jumping blocks is calculated based on the individual area of ​​each slag-jumping block, and the sum of the areas is determined as the slag-jumping area of ​​the image to be identified.

9. A slag splashing and furnace jumping identification device, characterized in that, The apparatus used in the slag splashing and furnace jumping information identification method according to any one of claims 1-8 includes: The information acquisition module is used to acquire the initial image and the image to be recognized; The model building module is used to train the initial network model based on the labeled initial images to obtain the scumbag recognition model; The image recognition module is used to input the image to be recognized into the scumbag recognition model and obtain the output result of the scumbag recognition model, the output result including multiple different labels; The image parsing module is used to assign different pixel values ​​to the image regions where different labels are located in the output results in order to obtain the target slag image; The debris jumping information determination module is used to identify the target debris jumping region in the target debris jumping image in order to obtain the debris jumping information of the image to be identified.

10. The slag splashing and furnace jumping identification device according to claim 9, characterized in that, The model building module includes: The image information recognition submodule is used to obtain slag jumping information based on the slag jumping image and to obtain background information based on the non-slag jumping image. The initial image includes the slag jumping image and the non-slag jumping image. An image cropping submodule is used to crop the initial image to obtain a sample image; The image annotation submodule is used to generate a first label based on the jump information, generate a second label based on the background information, and annotate the sample image based on the first label and the second label; The model training submodule is used to obtain a training sample dataset based on the labeled sample images, and to train the initial network model based on the training sample dataset to obtain the scumbag identification model.

11. The slag splashing and furnace jumping identification device according to claim 9, characterized in that, The module for determining "scumbag jumping" information includes: An image receiving submodule is used to receive the output image of the scumbag identification model, the output image including a preset first label and a preset second label; The image parsing submodule is used to assign a first pixel value to the area where the first label is located and a second pixel value to the area where the second label is located, so as to obtain the target slag image; The slag-jumping region determination submodule is used to calculate the area of ​​each slag-jumping block to obtain a legal slag-jumping block, and to determine the target slag-jumping region of the image to be identified based on the legal slag-jumping blocks; The slag jumping information collection submodule is used to calculate the slag jumping area of ​​the target slag jumping region, count the number of legal slag jumping blocks, and record the slag jumping distribution of multiple slag jumping blocks.

12. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the electronic device to implement the slag splashing protection furnace jumping information identification method as described in any one of claims 1 to 8.

13. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by the computer's processor, causes the computer to perform the slag splashing protection furnace jumping information identification method as described in any one of claims 1 to 8.

Images