Method, device and equipment for detecting overflow of slime water and readable storage medium

Abstract

The invention provides a method, device and equipment for detecting overflow of slime water and a readable storage medium, and the method comprises the steps: carrying out the recognition of a marker region image through a preset slime water detection model, and obtaining the coverage degree of a preset marker covered by the slime water in a region where the preset marker is located, the coal slime water detection model is obtained by training and verifying a basic target detection model through a plurality of image samples, and the plurality of image samples are obtained by labeling marker coverage degrees and labeling coal slime water overflow conditions on a plurality of images; the plurality of images are obtained by shooting images of different angles of areas where different markers are located; and determining the overflow condition of the slime water according to the coverage degree of the preset marker covered by the slime water. Through the method, the effect of accurately and timely detecting the overflow of the slime water can be achieved.

Description

Technical Field

[0001] The present application relates to the field of target detection, and in particular, to a method, device, equipment and readable storage medium for detecting coal slime water overflow. Background Art

[0002] In the production process of coal preparation plants, coal slime water is a mixture of fine coal particles and water, which is highly corrosive and polluting. If coal slime water overflows, it may cause serious pollution to the environment, increase the cost of cleaning and treatment, and even affect the normal operation of production equipment. Therefore, timely and accurate detection of coal slime water overflow is crucial for safe production and environmental protection in coal preparation plants. Traditional methods for detecting coal slime water overflow mainly rely on manual inspections or simple physical sensors such as level meters or flow meters.

[0003] These methods have problems such as low detection accuracy, slow response speed, and susceptibility to environmental interference. Especially in the complex operating environment of coal preparation plants, it is difficult to achieve real-time and automated spillover monitoring.

[0004] Therefore, how to accurately and timely detect coal slime water overflow is a technical problem that needs to be solved. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a method for detecting coal slime water overflow. Through the technical solution of the embodiments of the present application, the effect of accurately and timely detecting coal slime water overflow can be achieved.

[0006] In the first aspect, an embodiment of the present application provides a method for detecting coal slime water overflow, comprising: identifying a marker area image through a preset coal slime water detection model, and obtaining the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying a basic target detection model through multiple image samples, the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by taking images of different angles of the area where different markers are located; according to the coverage degree of the coal slime water covering the preset marker, the coal slime water overflow situation is determined.

[0007] In the above embodiments of the present application, through image processing and deep learning technology, by setting markers in the area, using machine vision technology to detect the state of the markers, and judging whether the coal slime water has overflowed, it is possible to achieve the effect of accurately and timely detecting the overflow of coal slime water.

[0008] In some embodiments, before identifying the marker area image through a preset coal slime water detection model and obtaining the coverage degree of the preset marker by the coal slime water in the area where the preset marker is located, it also includes: obtaining images of the areas where different markers are located to obtain multiple images, wherein the multiple images include images of different markers at various angles; annotating the multiple images with the marker coverage degree and the coal slime water overflow situation to obtain multiple image samples; training and verifying the basic target detection model through multiple image samples to obtain a coal slime water detection model.

[0009] In the above embodiments of the present application, by setting markers in areas where coal slime water may overflow, and forming image samples by taking images at different angles and annotating them, the basic target detection model can be trained, and the obtained coal slime water detection model is used to accurately judge the overflow of coal slime water.

[0010] In some embodiments, images of areas where different markers are located are obtained to obtain multiple images, including: multiple markers are set in key areas of the coal preparation plant, where the key areas include: coal slurry water pipeline areas and storage tank areas. Images of the areas where multiple markers are located are collected from various angles by a high-definition industrial camera to obtain multiple images.

[0011] In the above-mentioned embodiments of the present application, by collecting images in key areas where coal slime water overflow may occur, training samples for training a coal slime water detection model can be formed, and the training of a basic target detection model can be realized. The obtained coal slime water detection model is used to accurately judge the overflow of coal slime water.

[0012] In some embodiments, the overflow of coal slurry water is determined based on the coverage degree of preset markers by coal slurry water, including: when it is determined that the coverage degree is greater than or equal to a preset threshold, it is determined that the coal slurry water has overflowed; when it is determined that the coverage degree is less than the preset threshold, it is determined that the coal slurry water has not overflowed.

[0013] In the above embodiments of the present application, the overflow of coal slime water is accurately analyzed by comparing the coverage degree of the marker and the threshold.

[0014] In some embodiments, when it is determined that the coverage degree is greater than or equal to a preset threshold, after determining the coal slime water overflow, it also includes: triggering an overflow alarm and saving the marker area image; analyzing the time, location and coverage of the preset markers of the coal slime water overflow in the marker area image.

[0015] In the above-mentioned embodiments of the present application, when coal slime water overflows, an alarm can be promptly issued and relevant coal slime water overflow information can be analyzed, so that the staff can handle it in a timely manner according to the time, location and coverage of preset markers of the coal slime water overflow.

[0016] In some embodiments, before identifying the marker area image through a preset coal slurry water detection model and obtaining the coverage degree of the preset marker by the coal slurry water in the area where the preset marker is located, it also includes: preprocessing the collected initial marker area image to obtain the marker area image, wherein the preprocessing includes: at least one of grayscale, denoising and contrast enhancement.

[0017] In the above-mentioned embodiment of the present application, the accuracy of the coal slime water detection model can be improved by training the basic target detection model through the marker area image obtained after preprocessing the marker area image.

[0018] In some embodiments, the marker area image is identified by a preset coal slurry water detection model to obtain the coverage degree of the preset marker by the coal slurry water in the area where the preset marker is located, including: identifying the identification parameters of the preset marker in the marker area image by the coal slurry water detection model, wherein the identification parameters include: at least one of shape, specification and area; comparing the identification parameters with the initial parameters of the preset marker to obtain the coverage degree.

[0019] In the above embodiments of the present application, the coverage degree of the preset markers can be accurately determined by comparing the shape, specification and area changes of the preset markers.

[0020] In a second aspect, an embodiment of the present application provides a device for detecting coal slime water overflow, comprising:

[0021] An identification module is used to identify the marker area image through a preset coal slime water detection model, and obtain the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by taking images of different angles of the area where different markers are located;

[0022] The overflow detection module is used to determine the overflow of coal slime water according to the coverage degree of preset markers covered by the coal slime water.

[0023] Optionally, the device further comprises:

[0024] A training module is used for the recognition module to recognize the marker area image through a preset coal slime water detection model, and obtain images of different marker areas before obtaining the coverage degree of the preset marker covered by coal slime water in the area where the preset marker is located, so as to obtain multiple images, wherein the multiple images include images of different angles of different markers;

[0025] Marking the coverage of markers and the overflow of coal slime water on multiple images to obtain multiple image samples;

[0026] The basic target detection model is trained and verified through multiple image samples to obtain the coal slime water detection model.

[0027] Optionally, the training module is specifically used to:

[0028] Multiple markers are set up in key areas of the coal preparation plant, including the coal slurry water pipeline area and the storage tank area.

[0029] Through high-definition industrial cameras, multiple images are obtained by collecting images from various angles of the area where multiple markers are located.

[0030] Optionally, the overflow detection module is specifically used for:

[0031] When it is determined that the coverage degree is greater than or equal to a preset threshold, it is determined that the coal slurry water has overflowed;

[0032] When it is determined that the coverage degree is less than a preset threshold, it is determined that the coal slurry water has not overflowed.

[0033] Optionally, the device further comprises:

[0034] An alarm module, which is used for triggering an overflow alarm and saving a marker area image after the overflow detection module determines that the coverage degree is greater than or equal to a preset threshold and that the coal slime water has overflowed;

[0035] Analyze the time, location and coverage of preset markers of coal slime water overflow in the marker area image.

[0036] Optionally, the device further comprises:

[0037] A preprocessing module is used for the recognition module to recognize the marker area image through a preset coal slurry water detection model, and before obtaining the coverage degree of the preset marker covered by the coal slurry water in the area where the preset marker is located, to preprocess the collected initial marker area image to obtain the marker area image, wherein the preprocessing includes: at least one of graying, denoising and contrast enhancement.

[0038] Optionally, the identification module is specifically used for:

[0039] Identify identification parameters of preset markers in the marker area image by using the coal slime water detection model, wherein the identification parameters include: at least one of shape, specification and area;

[0040] The recognition parameters are compared with the initial parameters of the preset markers to obtain the coverage degree.

[0041] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method provided in the first aspect are performed.

[0042] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the method provided in the first aspect are performed.

[0043] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or be understood by practicing the embodiments of the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A flow chart of a method for detecting coal slime water overflow provided in an embodiment of the present application;

[0046] Figure 2 A flow chart of a method for training a coal slime water detection model provided in an embodiment of the present application;

[0047] Figure 3 A schematic block diagram of a device for detecting coal slime water overflow provided in an embodiment of the present application;

[0048] Figure 4 A schematic diagram of the structure of a device for detecting coal slime water overflow provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0050] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0051] First, some terms involved in the embodiments of the present application are explained to facilitate understanding by those skilled in the art.

[0052] The YOLOv8 model is the latest version of the YOLO (You Only Look Once) object detection and image detection model developed by Ultralytics. YOLOv8 is regarded as a cutting-edge, state-of-the-art model that builds on the success of previous YOLO and introduces new features and improvements to further improve performance and flexibility. It can be trained on large datasets and is able to run on a variety of hardware platforms, from CPUs to GPUs. A key feature of YOLOv8 is its scalability, as it is designed as a framework that supports all previous YOLO versions, making it easy to switch between different versions and compare their performance.

[0053] The TensorRT model is an optimized deep learning model designed for high-performance inference on NVIDIA GPUs. TensorRT is a high-performance deep learning inference optimizer and runtime acceleration library designed to provide low-latency, high-throughput deployment inference for deep learning applications.

[0054] The PyTorch model is the result of neural network training and optimization, which includes the neural network skeleton and learned parameters. PyTorch is an open source deep learning framework that mainly implements automatic differentiation and introduces dynamic computational graphs to make model building more flexible.

[0055] This application is applied to the scenario of..., the specific scenario is

[0056] In the production process of coal preparation plants, coal slime water is a mixture of fine coal particles and water, which is highly corrosive and polluting. If coal slime water overflows, it may cause serious pollution to the environment, increase the cost of cleaning and treatment, and even affect the normal operation of production equipment. Therefore, timely and accurate detection of coal slime water overflow is crucial for safe production and environmental protection in coal preparation plants. Traditional coal slime water overflow detection methods mainly rely on manual inspections or simple physical sensors such as liquid level meters or flow meters. These methods have problems such as low detection accuracy, slow response speed, and susceptibility to environmental interference. Especially in the complex operating environment of coal preparation plants, it is difficult to achieve real-time and automated overflow monitoring.

[0057] To this end, the present application uses a preset coal slime water detection model to identify the marker area image, and obtains the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, and the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by shooting images of different angles of the area where different markers are located; according to the coverage degree of the preset marker covered by the coal slime water, the overflow of the coal slime water is determined. Through image processing and deep learning technology, by setting markers in the area, using machine vision technology to detect the state of the marker, and judging whether the coal slime water has overflowed, the effect of accurately and timely detecting the overflow of coal slime water can be achieved.

[0058] In the embodiment of the present application, the execution entity may be a coal slime water overflow detection device in a coal slime water overflow detection system. In actual applications, the coal slime water overflow detection device may be an electronic device such as a terminal device and a server, which is not limited here.

[0059] Combine the following Figure 1 The method for detecting coal slime water overflow in an embodiment of the present application is described in detail.

[0060] Please see Figure 1 , Figure 1 A flow chart of a method for detecting coal slime water overflow provided in an embodiment of the present application is as follows: Figure 1 The methods shown for detecting coal slurry water spills include:

[0061] Step 110: Identify the marker area image using a preset coal slime water detection model to obtain the coverage degree of the preset marker by the coal slime water in the area where the preset marker is located.

[0062] Among them, the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, multiple image samples are obtained by annotating the coverage degree of markers and the coal slime water overflow conditions of multiple images, and multiple images are obtained by taking images from various angles of the areas where different markers are located.

[0063] In some embodiments of the present application, before the marker area image is identified by a preset coal slime water detection model to obtain the coverage degree of the preset marker by the coal slime water in the area where the preset marker is located, Figure 1 The method shown also includes: acquiring images of areas where different markers are located to obtain multiple images, wherein the multiple images include images of different markers at various angles; annotating the multiple images with respect to the degree of marker coverage and the coal slime water overflow conditions to obtain multiple image samples; and training and verifying the basic target detection model through multiple image samples to obtain a coal slime water detection model.

[0064] In the above process, the present application sets markers in areas where coal slime water may overflow, and forms image samples by taking images at different angles and annotating them, so as to realize the training of the basic target detection model. The obtained coal slime water detection model is used to accurately judge the overflow of coal slime water.

[0065] The marker may be a set marker object or a marker horizontal line. The multiple images include three types of images: full coverage, half coverage, and no coverage of the marker, and images without markers may also be mixed. The coverage degree indicates the percentage of the marker covered by the coal slime water. The overflow situation includes two situations: coal slime water overflow and no overflow. The basic target detection model may use YOLOv8.

[0066] Optionally, the OLOv8 model uses a standard convolution layer by default, but near the input layer of the model, the size of the feature map is large and the computational complexity of the standard convolution is high. Therefore, in order to reduce computational overhead and improve the inference speed of the model, this application replaces the first 5 layers of standard convolution near the input layer with depthwise separable convolution, which decomposes the standard convolution into depthwise convolution and pointwise convolution. This method can significantly reduce the computational complexity while maintaining the expressiveness of the convolution layer.

[0067] Furthermore, the replaced model is trained in a standard way, using part of the samples in the image sample for training, keeping the hyperparameters (such as initial learning rate, batch size, loss function, etc.) consistent with the standard YOLOv8 model, and using the validation set (another part of the samples in the image sample) to evaluate the model's precision, recall, and score, and recording these indicators to evaluate the model's baseline performance. Then, by analyzing the model weights and weight gradients, the importance of the convolution kernel is evaluated (the convolution kernel with a smaller absolute value of the weight usually contributes less to the final output of the model, and the convolution kernel with a smaller gradient value has a weaker effect on the model's learning), and the convolution kernels that contribute less to the model performance (do not meet the requirements) are cut off to reduce the complexity of the model.

[0068] Finally, the pruned model is retrained to fix the accuracy problems caused by pruning. When retraining, hyperparameters such as the learning rate are appropriately adjusted to ensure that the model can reach convergence again. All weights and activation values ​​are converted to sixteen-bit binary (FP16) representation, and the model is retrained under FP16 precision. Parameters such as the learning rate and the number of training rounds are tuned to adapt to the numerical instability that may be introduced by FP16 precision; after completing model optimization and retraining, the optimized model is used to input the latest images in the test set or actual scene into the model for inference to obtain the bounding box, confidence, and category information of the target detection. At the same time, the real-time performance (the number of images inferred per unit time) and resource consumption (GPU usage) of the model are evaluated. At the same time, the optimized model is comprehensively evaluated through indicators such as precision, recall, and score to obtain the final coal slime water detection model.

[0069] Optionally, in order to ensure the performance of the system in real-time and high-density traffic conditions, use TensorRT for deep learning reasoning acceleration. Use the TensorRT high-performance library interface to convert the trained coal slime water detection model (YOLOV8) from the PyTorch model to TensorRT to improve the model's reasoning speed and reduce resource consumption. And adopt a multi-threaded processing architecture to assign tasks such as image acquisition, preprocessing, target detection and tracking to different threads, execute them in parallel, and make full use of multi-core CPU and GPU resources.

[0070] In some embodiments of the present application, images of areas where different markers are located are obtained to obtain multiple images, including: multiple markers are set in key areas of the coal preparation plant, where the key areas include: coal slurry water pipeline areas and storage tank areas. Images of the areas where multiple markers are located are collected from various angles by a high-definition industrial camera to obtain multiple images.

[0071] In the above process, the present application can form training samples for training the coal slime water detection model by collecting images in key areas where coal slime water overflow may occur, and can realize the training of the basic target detection model. The obtained coal slime water detection model is used to accurately judge the overflow of coal slime water.

[0072] Among them, the key area can also be the production area where coal slime water is produced. The camera can be waterproof, dustproof and shockproof to adapt to the harsh working environment of the coal preparation plant. The camera installation angle should take into account the full coverage of the markers, avoid blind spots, and ensure that the markers are clearly visible in the image.

[0073] In some embodiments of the present application, before the marker area image is identified by a preset coal slime water detection model to obtain the coverage degree of the preset marker by the coal slime water in the area where the preset marker is located, Figure 1 The method shown also includes: preprocessing the acquired initial marker area image to obtain a marker area image, wherein the preprocessing includes: at least one of grayscale, denoising and contrast enhancement.

[0074] In the above process, the present application trains a basic target detection model by preprocessing the marker area image obtained by the marker area image, so as to improve the accuracy of the coal slime water detection model.

[0075] The preprocessing may also include operations such as brightness adjustment and cropping.

[0076] In some embodiments of the present application, the marker area image is identified by a preset coal slurry water detection model to obtain the coverage degree of the preset marker covered by the coal slurry water in the area where the preset marker is located, including: identifying the identification parameters of the preset marker in the marker area image by the coal slurry water detection model, wherein the identification parameters include: at least one of shape, specification and area; comparing the identification parameters with the initial parameters of the preset marker to obtain the coverage degree.

[0077] In the above process, the present application can accurately judge the coverage degree of the preset markers by comparing the changes in shape, specification and area of ​​the preset markers.

[0078] The initial parameters include the initial shape, specification and area of ​​the preset marker, etc. The recognition parameters may also include parameters such as color and contrast.

[0079] Step 120: Determine the overflow of the coal slurry water according to the coverage degree of the preset markers by the coal slurry water.

[0080] In some embodiments of the present application, the overflow of coal slurry water is determined based on the coverage degree of the preset markers covered by the coal slurry water, including: when it is determined that the coverage degree is greater than or equal to a preset threshold, it is determined that the coal slurry water has overflowed; when it is determined that the coverage degree is less than the preset threshold, it is determined that the coal slurry water has not overflowed.

[0081] In the above process, the present application accurately analyzes the overflow of coal slime water by comparing the coverage degree and threshold of the markers.

[0082] Among them, the preset threshold can be set according to needs.

[0083] In some embodiments of the present application, when it is determined that the coverage degree is greater than or equal to a preset threshold, after the coal slime water overflow is determined, Figure 1 The method shown also includes: triggering an overflow alarm and saving a marker area image; analyzing the time, location and coverage of preset markers of the coal slime water overflow in the marker area image.

[0084] In the above process, the present application can timely give an alarm and analyze relevant coal slime water overflow information when coal slime water overflows, so that the staff can handle it in time according to the time, location and coverage of preset markers of the coal slime water overflow.

[0085] Optionally, when it is identified that the marker is covered by coal slime water and the coverage reaches the set threshold, the system immediately determines it as a spill event. This judgment is based on real-time monitoring and analysis of ground markers. Once a spill event is confirmed, the system will automatically activate the alarm function. The alarm signal can be prompted by visual (such as flashing warning lights) or sound (such as alarms). Image recording and preservation: The system automatically saves image data of the spill area, including images of covered markers. The saved images are used for real-time viewing on site and archived for subsequent analysis and accident handling. Notification and response: The system sends coal slime water spill alarm notifications to relevant operators through the connected control center and communication equipment.

[0086] In the above Figure 1In the process shown, the present application identifies the marker area image through a preset coal slime water detection model, and obtains the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, and the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by shooting images of different angles of the area where different markers are located; according to the coverage degree of the preset marker covered by the coal slime water, the overflow situation of the coal slime water is determined. Through image processing and deep learning technology, by setting markers in the area, using machine vision technology to detect the state of the markers, and judging whether the coal slime water has overflowed, the effect of accurately and timely detecting the overflow of coal slime water can be achieved.

[0087] Combine the following Figure 2 The implementation method of detecting coal slime water overflow in the embodiment of the present application is described in detail.

[0088] Please see Figure 2 , Figure 2 A flow chart of a method for training a coal slime water detection model provided in an embodiment of the present application, such as Figure 2 The method for training the coal slime water detection model shown includes:

[0089] Step 210: Convolution replacement.

[0090] Specifically: Some standard convolutional layers of the basic object detection model are replaced with separable convolutional layers to obtain the initial image detection model.

[0091] Step 220: Initial model training.

[0092] Specifically: perform standard training on the replaced model, and use part of the image samples for training.

[0093] Step 230: Initial model evaluation.

[0094] Specifically: Use the validation set (another part of the image samples) to evaluate the model's precision, recall, and score, and record these indicators to evaluate the baseline performance of the model.

[0095] Step 240: Model optimization.

[0096] Specifically: By analyzing the model weights and weight gradients, the importance of the convolution kernels is evaluated, and the convolution kernels that contribute little to the model performance (do not meet the requirements) are pruned to reduce the complexity of the model.

[0097] Step 250: Model hybrid training.

[0098] Specifically: Retrain the pruned model to fix the accuracy issues caused by pruning. When retraining, adjust hyperparameters such as learning rate appropriately to ensure that the model can reach convergence again.

[0099] Step 260: Model inference.

[0100] Specifically: Use the optimized model to input the latest images in the test set or actual scene into the model, perform inference, and obtain the bounding box, confidence, and category information of the target detection.

[0101] Step 270: Model performance and accuracy evaluation.

[0102] Specifically: evaluate the real-time performance (the number of inference images per unit time) and resource consumption (GPU usage rate) of the model, and comprehensively evaluate the optimized model through indicators such as precision, recall rate, and score to obtain the final coal slime water detection model.

[0103] also, Figure 2 The specific methods and steps shown can be found in Figure 1 The method shown will not be described in detail here.

[0104] Previous article Figure 1-Figure 2 Describes the method of detecting coal slime water overflow, the following is combined with Figure 3-Figure 4 Describe the apparatus for detecting overflow of coal slurry water.

[0105] Please refer to Figure 3 , is a schematic block diagram of a device 300 for detecting coal slime water overflow provided in an embodiment of the present application, and the device 300 may be a module, program segment or code on an electronic device. Figure 1 The method embodiment corresponds to and can be executed Figure 1 The various steps involved in the method embodiment and the specific functions of the device 300 can be found in the description below. To avoid repetition, the detailed description is appropriately omitted here.

[0106] Optionally, the device 300 includes:

[0107] The recognition module 310 is used to recognize the marker area image through a preset coal slime water detection model, and obtain the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by taking images of different angles of the area where different markers are located;

[0108] The overflow detection module 320 is used to determine the overflow of the coal slime water according to the coverage degree of the preset markers covered by the coal slime water.

[0109] Optionally, the device further comprises:

[0110] A training module is used for the recognition module to recognize the marker area image through a preset coal slime water detection model, and before obtaining the coverage degree of coal slime water in the area where the preset marker is located, obtain images of the area where different markers are located to obtain multiple images, wherein the multiple images include images of different markers at various angles; annotate the multiple images with the marker coverage degree and the coal slime water overflow situation to obtain multiple image samples; train and verify the basic target detection model through multiple image samples to obtain a coal slime water detection model.

[0111] Optionally, the training module is specifically used to:

[0112] Multiple markers are set up in key areas of the coal preparation plant, including the coal slurry water pipeline area and the storage tank area. Through high-definition industrial cameras, multiple images are collected from various angles of the areas where multiple markers are located to obtain multiple images.

[0113] Optionally, the overflow detection module is specifically used for:

[0114] When it is determined that the coverage degree is greater than or equal to a preset threshold, it is determined that the coal slurry water has overflowed; when it is determined that the coverage degree is less than the preset threshold, it is determined that the coal slurry water has not overflowed.

[0115] Optionally, the device further comprises:

[0116] The alarm module is used for triggering the overflow alarm and saving the marker area image after the overflow detection module determines that the coverage degree is greater than or equal to the preset threshold value and the coal slime water overflows; analyzing the time, location and coverage of the preset markers in the marker area image.

[0117] Optionally, the device further comprises:

[0118] A preprocessing module is used for the recognition module to recognize the marker area image through a preset coal slurry water detection model, and before obtaining the coverage degree of the preset marker covered by the coal slurry water in the area where the preset marker is located, to preprocess the collected initial marker area image to obtain the marker area image, wherein the preprocessing includes: at least one of graying, denoising and contrast enhancement.

[0119] Optionally, the identification module is specifically used for:

[0120] The recognition parameters of the preset markers in the marker area image are recognized by the coal slime water detection model, wherein the recognition parameters include: at least one of shape, specification and area; the recognition parameters are compared with the initial parameters of the preset markers to obtain the coverage degree.

[0121] Please refer to Figure 4 4 is a schematic block diagram of a device for detecting coal slime water overflow provided in an embodiment of the present application. The device may include a memory 410 and a processor 420. Optionally, the device may also include: a communication interface 430 and a communication bus 440. The device is similar to the above Figure 1 The method embodiment corresponds to and can be executed Figure 1 The various steps involved in the method embodiment and the specific functions of the device can be found in the description below.

[0122] Specifically, the memory 410 is used to store computer-readable instructions.

[0123] Processor 420 is used to process the readable instructions stored in the memory and can execute Figure 1 The steps in the method.

[0124] The communication interface 430 is used for signaling or data communication with other node devices, for example, for communication with a server or a terminal, or for communication with other device nodes, but the embodiments of the present application are not limited thereto.

[0125] The communication bus 440 is used to realize direct connection and communication among the above components.

[0126] The communication interface 430 of the device in the embodiment of the present application is used to communicate signals or data with other node devices. The memory 410 can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The memory 410 can also be at least one storage device located away from the aforementioned processor. The memory 410 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 420, the electronic device executes the aforementioned Figure 1The method process shown. The processor 420 can be used on the device 300 and is used to perform the functions in the present application. Exemplarily, the above-mentioned processor 420 can be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and the embodiments of the present application are not limited thereto.

[0127] The embodiment of the present application also provides a readable storage medium, when the computer program is executed by a processor, Figure 1 The method process in the method embodiment shown is executed by the electronic device.

[0128] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method, and will not be described in detail here.

[0129] In summary, the embodiments of the present application provide a method, device, equipment and readable storage medium for detecting coal slime water overflow, the method comprising: identifying the marker area image through a preset coal slime water detection model, obtaining the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by taking images of different angles of the area where different markers are located; according to the coverage degree of the preset marker covered by the coal slime water, the coal slime water overflow situation is determined. This method can achieve the effect of accurately and timely detecting coal slime water overflow.

[0130] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0131] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0132] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several 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 methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0133] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0134] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0135] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

Claims

1. A method for detecting coal slime water overflow, characterized in that: include: The marker area image is identified by a preset coal slime water detection model to obtain the coverage degree of the preset marker by coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by taking images of different angles of the area where different markers are located; The overflow of the coal slime water is determined according to the coverage degree of the preset marker by the coal slime water.

2. The method according to claim 1, characterized in that: Before identifying the marker area image by using the preset coal slime water detection model to obtain the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, the method further includes: Acquire images of the areas where the different markers are located to obtain the multiple images, wherein the multiple images include images of the different markers at various angles; Marking the multiple images with respect to the coverage degree of the markers and the overflow of the coal slime water, to obtain the multiple image samples; The basic target detection model is trained and verified through the multiple image samples to obtain the coal slime water detection model.

3. The method according to claim 2, characterized in that The method of acquiring images of areas where different markers are located to obtain multiple images includes: A plurality of markers are set in key areas of the coal preparation plant, wherein the key areas include: a coal slurry water pipeline area and a storage tank area; The multiple images are obtained by collecting images of various angles of the area where the multiple markers are located using a high-definition industrial camera.

4. The method according to any one of claims 1 to 3, characterized in that: The determining the overflow of the coal slime water according to the coverage degree of the preset marker by the coal slime water includes: When it is determined that the coverage degree is greater than or equal to a preset threshold, it is determined that the coal slime water has overflowed; When it is determined that the coverage degree is less than the preset threshold, it is determined that the coal slurry water has not overflowed.

5. The method according to claim 4, characterized in that When it is determined that the coverage degree is greater than or equal to a preset threshold, after determining that the coal slime water has overflowed, the method further includes: triggering an overflow alarm and saving an image of the marker area; The time and location of the coal slime water overflow in the marker area image and the coverage of the preset markers are analyzed.

6. The method according to any one of claims 1 to 3, characterized in that: Before identifying the marker area image by using the preset coal slime water detection model to obtain the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, the method further includes: The collected initial marker region image is preprocessed to obtain the marker region image, wherein the preprocessing includes: at least one of graying, denoising and contrast enhancement.

7. The method according to any one of claims 1 to 3, characterized in that: The identifying the marker area image by using a preset coal slime water detection model to obtain the coverage degree of the preset marker by the coal slime water in the area where the preset marker is located includes: Identifying identification parameters of the preset marker in the marker region image by using the coal slime water detection model, wherein the identification parameters include: at least one of shape, specification and area; The recognition parameters are compared with the initial parameters of the preset markers to obtain the coverage degree.

8. A device for detecting coal slime water overflow, characterized in that: include: An identification module is used to identify the marker area image through a preset coal slime water detection model, and obtain the coverage degree of the preset marker covered by the coal slime water in the area where the preset marker is located, wherein the coal slime water detection model is obtained by training and verifying the basic target detection model through multiple image samples, and the multiple image samples are obtained by annotating the marker coverage degree and the coal slime water overflow situation of multiple images, and the multiple images are obtained by taking images of different angles of the area where different markers are located; The overflow detection module is used to determine the overflow of the coal slime water according to the coverage degree of the preset marker by the coal slime water.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1 to 7 are executed.

10. A computer-readable storage medium, characterized in that: include: A computer program, when the computer program is run on a computer, causes the computer to execute the method according to any one of claims 1 to 7.

Images