Water leakage identification method and related equipment

Abstract

The invention provides a water leakage identification method and related equipment. The method comprises the following steps: acquiring an image at a first pipeline to obtain a first image; identifying the first image by using a trained identification model to obtain a first result; responding to the condition that the first result is water leakage, continuously obtaining the image of the first pipeline, and obtaining at least one second image; identifying each second image by using the trained identification model to obtain at least one second result; and identifying the water leakage condition of the first pipeline based on the second result. According to the embodiment of the invention, the real-time inspection and analysis of the pipeline state are realized, the water leakage point is found and alarmed in time, and the response speed and accuracy of inspection are improved. Even if the signal of the underground powerhouse is not good, the defect judgment can be made in the equipment, and the integrity of the inspection range is kept.

Description

Technical Field

[0001] The present application relates to the technical field of pipeline leakage identification, and in particular to a leakage identification method and related equipment. Background Art

[0002] In the field of pipeline leakage inspection, the existing technical solutions mainly include the following categories from the perspective of identification methods:

[0003] Methods based on visual inspection: rely on naked eye observation to identify leaks, such as water stains at pipe joints, water marks on walls, or mildew on ceilings. However, this method relies on manual inspections, which is inefficient, and the detection accuracy is affected by the experience of the inspectors.

[0004] Pressure test-based method: Water leaks are detected by injecting air or water into the water pipe system and monitoring the pressure changes after pressurization. If the pressure drops rapidly, there may be a leak. However, this method requires shutting down the water system to be tested and isolating it, which cannot be tested during the operation of the pumped storage power station water system; and it can usually only determine whether the pipeline is leaking, but cannot determine the specific leak point.

[0005] Gas tracer-based method: By injecting a harmless gas (such as a mixture of hydrogen and nitrogen) into the pipeline, and then using a special gas detector to detect gas leakage around the pipeline, the leak point is determined. However, this method requires the water system to be shut down and isolated during testing, and cannot be tested during the operation of the pumped storage power station water system.

[0006] Voiceprint-based method: The location and amount of water leakage are determined by picking up the sound of water leakage. When a pipe leaks, the water flow noise will change, and the leak point can be located by analyzing these noises. However, this method is greatly affected by external environmental noise and may be difficult to detect accurately in a noisy environment.

[0007] Thermal imaging-based method: Use infrared thermal imagers to detect temperature changes caused by water leakage. Water pipe leakage will cause changes in the temperature of surrounding materials, and these temperature differences can be intuitively seen through thermal imagers. However, the imaging effect of this method may be affected by the material type and ambient temperature.

[0008] Flow-based method: Use electronic sensors or detectors to monitor changes in water flow in pipes. If abnormal flow is found, there may be a leak. However, this method is difficult to locate the specific leak point. Summary of the invention

[0009] In view of this, the purpose of this application is to provide a water leakage identification method and related equipment.

[0010] Based on the above purpose, the present application provides a water leakage identification method, comprising:

[0011] Acquire an image at a first pipeline to obtain a first image;

[0012] Using the trained recognition model to recognize the first image, to obtain a first result;

[0013] In response to the first result being water leakage, continuing to acquire images at the first pipeline to obtain at least one second image;

[0014] Using the trained recognition model to recognize each of the second images, to obtain at least one second result;

[0015] The water leakage of the first pipeline is identified based on the second result.

[0016] In a possible implementation, the trained recognition model is trained by the following method:

[0017] Constructing a recognition model to be trained; the recognition model to be trained includes a backbone network, a neck network and a head network;

[0018] Get pipeline image;

[0019] Extracting a first feature in the pipeline image using the backbone network;

[0020] Using the neck network to perform feature fusion on the first feature to obtain a second feature;

[0021] Using the head network to detect the second feature, and obtain a first result for training;

[0022] Based on the first training result and the real result corresponding to the pipeline image, the recognition model to be trained is trained to obtain the trained recognition model.

[0023] In a possible implementation, the backbone network includes a first convolutional layer, a second convolutional layer, a third convolutional layer, a fourth convolutional layer, and a fifth convolutional layer;

[0024] The extracting the first feature in the pipeline image by using the backbone network includes:

[0025] extracting a first image feature in the pipeline image using the first convolutional layer;

[0026] Using the second convolutional layer to perform convolution and fusion processing on the first image feature to obtain a second image feature;

[0027] Using the third convolutional layer to perform convolution and fusion processing on the second image feature to obtain a third image feature;

[0028] Using the fourth convolutional layer to perform convolution and fusion processing on the third image feature to obtain a fourth image feature;

[0029] The fourth image feature is fused and enhanced by using the fifth convolutional layer to obtain the first feature.

[0030] In a possible implementation, the using the neck network to perform feature fusion on the first feature to obtain the second feature includes:

[0031] Performing upsampling processing on the first feature and performing feature fusion processing on the fourth image feature to obtain a first fusion result;

[0032] The first fusion result is upsampled and fused with the third image feature to obtain the second feature.

[0033] In a possible implementation manner, identifying the water leakage of the first pipeline based on the second result includes:

[0034] Set the number of tokens in the preset token bucket to zero;

[0035] determining the number of tokens based on the second result;

[0036] In response to the number of tokens reaching a preset threshold, recording the condition of the first pipeline as a water leak;

[0037] In response to the number of tokens being less than the preset threshold, the number of tokens is set to zero.

[0038] In a possible implementation manner, determining the number of tokens based on the second result includes:

[0039] In response to the second result being a water leak, increasing the number of tokens by a first amount;

[0040] In response to the second result being that there is no water leakage, the number of tokens is reduced by a second amount.

[0041] Based on the same inventive concept, the embodiment of the present application also provides a water leakage identification device, including:

[0042] A first acquisition module is configured to acquire an image at a first pipeline to obtain a first image;

[0043] A first recognition module is configured to recognize the first image using a trained recognition model to obtain a first result;

[0044] A second acquisition module is configured to, in response to the first result being water leakage, continue to acquire the image at the first pipeline to obtain at least one second image;

[0045] A second recognition module is configured to recognize each of the second images using the trained recognition model to obtain at least one second result;

[0046] The third identification module is configured to identify the water leakage of the first pipeline based on the second result.

[0047] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a water leakage identification method as described in any one of the above items is implemented.

[0048] Based on the same inventive concept, an embodiment of the present application further provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute any of the above-mentioned water leakage identification methods.

[0049] Based on the same inventive concept, an embodiment of the present application further provides a computer program product, which includes computer program instructions, and the computer instructions are used to enable the computer program product to execute any of the above-mentioned water leakage identification methods.

[0050] As can be seen from the above, the water leakage identification method and related equipment provided by the present application obtain a first image by acquiring an image at a first pipeline; using a trained recognition model to identify the first image to obtain a first result; in response to the first result being a water leakage, continue to acquire the image at the first pipeline to obtain at least one second image; use the trained recognition model to identify each of the second images to obtain at least one second result; and identify the water leakage of the first pipeline based on the second result. The embodiment of the present application can quickly cover a large area for pipeline leakage detection by using a fixed high-definition camera and an algorithm model carried by a drone. Compared with traditional manual inspections, the inspection time is greatly shortened and the inspection efficiency is improved. The use of fixed high-definition cameras and drone inspections reduces the dependence on human resources and reduces the cost of long-term operation. At the same time, the efficient performance of the algorithm reduces the dependence on high-performance computing resources and further reduces costs. The use of fixed high-definition cameras and drone inspections reduces the risk of personnel working at high altitudes and avoids safety accidents caused by human factors. The automated inspection process also reduces the error rate of human operations and ensures the safety of the inspection process. It realizes real-time inspection and analysis of pipeline status, timely discovers and alarms water leakage points, and improves the response speed and accuracy of inspection. Even when the signal in the underground plant is poor, it can make defect judgments in the equipment to maintain the integrity of the inspection range. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 A schematic diagram of a water leakage identification method according to an embodiment of the present application;

[0053] Figure 2 A detailed flow chart of the water leakage identification method according to an embodiment of the present application;

[0054] Figure 3 A schematic diagram of a pipeline water leakage neural network based on YOLOv8 in an embodiment of the present application;

[0055] Figure 4 A schematic diagram of a Detect neural network according to an embodiment of the present application;

[0056] Figure 5 This is a schematic diagram of the structure of a water leakage identification device according to an embodiment of the present application;

[0057] Figure 6This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0059] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0060] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0061] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0062] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0063] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0064] As mentioned in the background technology section, in the field of pipeline leakage inspection, the existing technical solutions mainly include the following categories from the perspective of identification methods:

[0065] Methods based on visual inspection: rely on naked eye observation to identify leaks, such as water stains at pipe joints, water marks on walls, or mildew on ceilings. However, this method relies on manual inspections, which is inefficient, and the detection accuracy is affected by the experience of the inspectors.

[0066] Pressure test-based method: Water leaks are detected by injecting air or water into the water pipe system and monitoring the pressure changes after pressurization. If the pressure drops rapidly, there may be a leak. However, this method requires shutting down the water system to be tested and isolating it, which cannot be tested during the operation of the pumped storage power station water system; and it can usually only determine whether the pipeline is leaking, but cannot determine the specific leak point.

[0067] Gas tracer-based method: By injecting a harmless gas (such as a mixture of hydrogen and nitrogen) into the pipeline, and then using a special gas detector to detect gas leakage around the pipeline, the leak point is determined. However, this method requires the water system to be shut down and isolated during testing, and cannot be tested during the operation of the pumped storage power station water system.

[0068] Voiceprint-based method: The location and amount of water leakage are determined by picking up the sound of water leakage. When a pipe leaks, the water flow noise will change, and the leak point can be located by analyzing these noises. However, this method is greatly affected by external environmental noise and may be difficult to detect accurately in a noisy environment.

[0069] Thermal imaging-based method: Use infrared thermal imagers to detect temperature changes caused by water leakage. Water pipe leakage will cause changes in the temperature of surrounding materials, and these temperature differences can be intuitively seen through thermal imagers. However, the imaging effect of this method may be affected by the material type and ambient temperature.

[0070] Flow-based method: Use electronic sensors or detectors to monitor changes in water flow in pipes. If abnormal flow is found, there may be a leak. However, this method is difficult to locate the specific leak point.

[0071] Based on the above considerations, the embodiment of the present application proposes a water leakage identification method, which obtains a first image by acquiring an image at a first pipeline; uses a trained recognition model to identify the first image to obtain a first result; in response to the first result being a water leakage, continues to acquire images at the first pipeline to obtain at least one second image; uses the trained recognition model to identify each of the second images to obtain at least one second result; and identifies the water leakage of the first pipeline based on the second result. The embodiment of the present application can quickly cover a large area for pipeline leakage detection by using a fixed high-definition camera and an algorithm model carried by a drone. Compared with traditional manual inspections, the inspection time is greatly shortened and the inspection efficiency is improved. The use of fixed high-definition cameras and drone inspections reduces the dependence on human resources and reduces the cost of long-term operation. At the same time, the efficient performance of the algorithm reduces the dependence on high-performance computing resources and further reduces costs. The use of fixed high-definition cameras and drone inspections reduces the risk of personnel working at high altitudes and avoids safety accidents caused by human factors. The automated inspection process also reduces the error rate of human operations and ensures the safety of the inspection process. It realizes real-time inspection and analysis of pipeline status, timely discovers and alarms water leakage points, and improves the response speed and accuracy of inspection. Even when the signal in the underground plant is poor, it can make defect judgments in the equipment to maintain the integrity of the inspection range.

[0072] The technical solutions of the embodiments of the present application are described in detail below through specific examples.

[0073] refer to Figure 1 The water leakage identification method of the embodiment of the present application comprises the following steps:

[0074] Step S101, acquiring an image at a first pipeline to obtain a first image;

[0075] Step S102, using the trained recognition model to recognize the first image to obtain a first result;

[0076] Step S103, in response to the first result being water leakage, continue to acquire the image of the first pipeline to obtain at least one second image;

[0077] Step S104, using the trained recognition model to recognize each of the second images to obtain at least one second result;

[0078] Step S105: identifying the water leakage of the first pipeline based on the second result.

[0079] refer to Figure 2 , is a detailed flow chart of the water leakage identification method according to an embodiment of the present application.

[0080] The following is based on Figure 1 and Figure 2 The embodiments of the present application are described in detail.

[0081] With respect to step S101, in the embodiment of the present application, a camera or a drone with a camera may be used to obtain an image of the first pipeline to obtain a first image. The first pipeline may be any position of the pipeline and may be a point to be detected.

[0082] With respect to step S102, the first image is recognized using the trained recognition model to obtain a first result.

[0083] In some embodiments, the trained recognition model is trained by the following method: constructing a recognition model to be trained; the recognition model to be trained includes a backbone network, a neck network and a head network; obtaining a pipeline image; using the backbone network to extract a first feature in the pipeline image; using the neck network to perform feature fusion on the first feature to obtain a second feature; using the head network to detect the second feature to obtain a first result for training; based on the first result for training and the real result corresponding to the pipeline image, training the recognition model to be trained to obtain the trained recognition model.

[0084] In the embodiment of the present application, this stage includes a total of three steps: data collection and preprocessing, and model training.

[0085] The first step is data collection and preprocessing.

[0086] This application first collects a large amount of image data about pipeline leakage in pumped storage power stations, and annotates and preprocesses these data in detail for subsequent model training. Specifically, this application uses 500 pictures of pipeline leakage, of which 400 are used as training sets for model learning; the other 100 are used as validation sets to evaluate the performance of the model. In the preprocessing, this application specifically marks the location of the water pipe leakage in the picture, and saves the label and the coordinates of the marked position in the document for model input.

[0087] Furthermore, the model needs to be trained.

[0088] Next, in an embodiment of the present application, the collected data is trained using an advanced YOLOv8 algorithm to generate a deep learning model that can accurately identify pipe leakage.

[0089] The Chinese name of the YOLOv8 algorithm is usually translated as "You Only Look Once (Eighth Edition)" or simply "YOLO Eighth Edition". YOLO is the abbreviation of "You Only Look Once", which means that this type of algorithm can complete the object detection task in one forward propagation. YOLOv8 is the eighth version of this series of algorithms, representing further optimization and improvement in accuracy, speed and other characteristics.

[0090] During the training process, this application uses a variety of optimization strategies, such as data enhancement and learning rate adjustment, to improve the accuracy and robustness of the model. The specific training parameters are set as follows: number of iterations epochs = 500, image size imgsz = 640, number of worker threads workers = 8, batch size batch = 16, learning rate lr = 0.01. After sufficient training, this application obtains a model with excellent performance.

[0091] refer to Figure 3 , is a schematic diagram of a pipeline water leakage neural network based on YOLOv8 in an embodiment of the present application.

[0092] like Figure 3 As shown in the figure, the network structure of the YOLOv8 algorithm mainly includes three modules: Backbone, Neck, and Head. The Backbone network is the basis of the model and is responsible for extracting features from the input pipeline leakage image. These features are the basis for target detection in subsequent network layers; the Neck network is located between the Backbone network and the Head network, and its function is to fuse and enhance the features of the pipeline leakage image. The Head network is the decision-making part of the target detection model and is responsible for generating the final detection results. In the scenario of pipeline leakage detection, the Head will determine whether there is a leakage problem based on the features provided by the Backbone and Neck, and may further determine the specific location of the leakage.

[0093] In the network structure of the model, the structure of the convolution module (Conv) contains three components in sequence: convolution layer (Conv2d), batch normalization (BatchNorm2d), and activation function (ReLU).

[0094] In some embodiments, the backbone network includes a first convolutional layer, a second convolutional layer, a third convolutional layer, a fourth convolutional layer and a fifth convolutional layer; the extracting the first feature in the pipeline image using the backbone network includes: extracting the first image feature in the pipeline image using the first convolutional layer; performing convolution and fusion processing on the first image feature using the second convolutional layer to obtain a second image feature; performing convolution and fusion processing on the second image feature using the third convolutional layer to obtain a third image feature; performing convolution and fusion processing on the third image feature using the fourth convolutional layer to obtain a fourth image feature; and performing fusion and enhancement processing on the fourth image feature using the fifth convolutional layer to obtain the first feature.

[0095] In the embodiment of the present application, the Backbone network includes five layers of convolutional modules (Conv): P1, P2, P3, P4, and P5.

[0096] P1: Conv module, convolution kernel size k = 3, stride s = 2, padding p = 1, used to extract image features.

[0097] P2, P3, and P4 are the combination of Conv module and C2f module. The convolution kernel size is k=3, the stride is s=2, and the padding is p=1. C2f is used to fuse the underlying feature map with the high-level feature map, thereby improving the accuracy of feature extraction and the operating efficiency of the model.

[0098] P5: Conv module, C2f module, SPPF module. Among them, the convolution kernel size k = 3, the stride s = 2, and the padding p = 1; the SPPF module is used to fuse and enhance features to improve the accuracy of target detection.

[0099] In some embodiments, the use of the neck network to perform feature fusion on the first feature to obtain the second feature includes: upsampling the first feature, performing feature fusion processing with the fourth image feature, and obtaining a first fusion result; upsampling the first fusion result, performing feature fusion processing with the third image feature, and obtaining the second feature.

[0100] In this embodiment, the Neck network mainly includes four modules: upsampling, C2f, Conv, and CONCAT.

[0101] After upsampling, the output features of the P5 layer are fused with the output features of the P4 layer through CONCAT and C2f. After upsampling, the fusion result Rm1 is fused with the output features of the P3 layer again through CONCAT and C2f modules. The output result R1 is input to the detector Detect in the Head network.

[0102] The output features of R1 after the convolution module are fused with the features of Rm1 through the CONCAT and C2f modules, and the output result R2 is input into the detector Detect in the Head network.

[0103] The output features of R2 after the convolution module are fused with the output features of the P5 layer through the CONCAT and C2f modules, and the output result R3 is input into the detector Detect in the Head network.

[0104] refer to Figure 4 , which is a schematic diagram of the Detect neural network of an embodiment of the present application.

[0105] The head network contains three detectors, the structure is as follows Figure 4 As shown in Figure 2. Each Detect consists of a detection head Bbox and a classification head Cls.

[0106] The detection head is responsible for predicting the bounding box regression value of each anchor box and the confidence of the target existence. It contains 3 layers of network: the first layer is a convolution module with a convolution kernel size of k=3, a stride of s=1, and a padding of p=1; the second layer is a convolution module with a convolution kernel size of k=3, a stride of s=1, and a padding of p=1; the third layer is a separate convolution layer (Conv2d) with a convolution kernel size of k=1, a stride of s=1, and a padding of p=0.

[0107] The classification head classifies each feature map and outputs the probability distribution of each category by reducing the dimension of the feature map. It contains 3 layers of network: the first layer is a convolution module with a convolution kernel size of k=3, a stride of s=1, and a padding of p=1; the second layer is a convolution module with a convolution kernel size of k=3, a stride of s=1, and a padding of p=1; the third layer is a separate convolution layer (Conv2d) with a convolution kernel size of k=1, a stride of s=1, and a padding of p=0.

[0108] Furthermore, the model is deployed. The trained YOLOv8 model is deployed on a camera or a drone with a camera, so that it can detect water leakage in the pipeline of the pumped storage power station in real time. At the same time, the token bucket algorithm is combined to further judge the detected defective images.

[0109] For steps S103-105, in response to the first result being water leakage, continue to acquire images of the first pipeline to obtain at least one second image; use the trained recognition model to recognize each of the second images to obtain at least one second result; and identify the water leakage of the first pipeline based on the second result.

[0110] In some embodiments, the identifying of the water leakage of the first pipeline based on the second result includes: setting the number of tokens in a preset token bucket to zero; determining the number of tokens based on the second result; in response to the number of tokens reaching a preset threshold, recording the condition of the first pipeline as a water leakage; in response to the number of tokens being less than the preset threshold, setting the number of tokens to zero.

[0111] In some embodiments, determining the number of tokens based on the second result includes: in response to the second result being a water leak, increasing the number of tokens by a first amount; in response to the second result being no water leak, decreasing the number of tokens by a second amount.

[0112] The token bucket algorithm in the embodiment of the present application is as follows: when a picture with defects is collected, the number of tokens in the token bucket is increased by 8; if no defects are found in the subsequent recorded pictures, the number of tokens is reduced by 2. When the number of tokens reaches 100, a defect is reported; when the number of tokens does not reach 100 after the analysis of 20 pictures is completed, the counter is cleared to 0 and no defect is reported. The token bucket method can reduce the misjudgment rate of the model and increase accuracy.

[0113] It can be seen from the above embodiments that the water leakage identification method described in the embodiment of the present application obtains a first image by acquiring an image at a first pipeline; uses a trained recognition model to identify the first image to obtain a first result; in response to the first result being a water leakage, continues to acquire images at the first pipeline to obtain at least one second image; uses the trained recognition model to identify each of the second images to obtain at least one second result; and identifies the water leakage of the first pipeline based on the second result. The embodiment of the present application can quickly cover a large area for pipeline leakage detection by using a fixed high-definition camera and an algorithm model carried by a drone. Compared with traditional manual inspections, the inspection time is greatly shortened and the inspection efficiency is improved. The use of fixed high-definition cameras and drone inspections reduces the dependence on human resources and reduces the cost of long-term operation. At the same time, the efficient performance of the algorithm reduces the dependence on high-performance computing resources and further reduces costs. The use of fixed high-definition cameras and drone inspections reduces the risk of personnel working at high altitudes and avoids safety accidents caused by human factors. The automated inspection process also reduces the error rate of human operations and ensures the safety of the inspection process. It realizes real-time inspection and analysis of pipeline status, timely discovers and alarms water leakage points, and improves the response speed and accuracy of inspection. Even when the signal in the underground plant is poor, it can make defect judgments in the equipment to maintain the integrity of the inspection range.

[0114] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0115] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0116] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a water leakage identification device.

[0117] refer to Figure 5 , the water leakage identification device comprises:

[0118] A first acquisition module is configured to acquire an image at a first pipeline to obtain a first image;

[0119] A first recognition module is configured to recognize the first image using a trained recognition model to obtain a first result;

[0120] A second acquisition module is configured to, in response to the first result being water leakage, continue to acquire the image at the first pipeline to obtain at least one second image;

[0121] A second recognition module is configured to recognize each of the second images using the trained recognition model to obtain at least one second result;

[0122] The third identification module is configured to identify the water leakage of the first pipeline based on the second result.

[0123] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0124] The device of the above embodiment is used to implement the corresponding water leakage identification method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0125] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the water leakage identification method described in any of the above embodiments is implemented.

[0126] Figure 6 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0127] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0128] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0129] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0130] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0131] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0132] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0133] The electronic device of the above embodiment is used to implement the corresponding water leakage identification method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0134] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the water leakage identification method described in any of the above embodiments.

[0135] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0136] The computer instructions stored in the storage medium of the above embodiments are used to enable the computer to execute the water leakage identification method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0137] Based on the same inventive concept, corresponding to the water leakage identification method described in any of the above embodiments, the present disclosure further provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor execute the water leakage identification method. Corresponding to the execution subject corresponding to each step in each embodiment of the water leakage identification method, the processor that executes the corresponding step can belong to the corresponding execution subject.

[0138] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the water leakage identification method described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0139] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0140] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0141] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0142] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A water leakage identification method, characterized in that: include: Acquire an image at a first pipeline to obtain a first image; Using the trained recognition model to recognize the first image, to obtain a first result; In response to the first result being water leakage, continuing to acquire images at the first pipeline to obtain at least one second image; Using the trained recognition model to recognize each of the second images, to obtain at least one second result; The water leakage of the first pipeline is identified based on the second result.

2. The method according to claim 1, characterized in that The trained recognition model is trained by the following method: Constructing a recognition model to be trained; the recognition model to be trained includes a backbone network, a neck network and a head network; Get pipeline image; Extracting a first feature in the pipeline image using the backbone network; Using the neck network to perform feature fusion on the first feature to obtain a second feature; Using the head network to detect the second feature, and obtain a first result for training; Based on the first training result and the real result corresponding to the pipeline image, the recognition model to be trained is trained to obtain the trained recognition model.

3. The method according to claim 2, characterized in that The backbone network includes a first convolutional layer, a second convolutional layer, a third convolutional layer, a fourth convolutional layer and a fifth convolutional layer; The extracting the first feature in the pipeline image by using the backbone network includes: extracting a first image feature in the pipeline image using the first convolutional layer; Using the second convolutional layer to perform convolution and fusion processing on the first image feature to obtain a second image feature; Using the third convolutional layer to perform convolution and fusion processing on the second image feature to obtain a third image feature; Using the fourth convolutional layer to perform convolution and fusion processing on the third image feature to obtain a fourth image feature; The fourth image feature is fused and enhanced by using the fifth convolutional layer to obtain the first feature.

4. The method according to claim 3, characterized in that The step of using the neck network to perform feature fusion on the first feature to obtain a second feature includes: Performing upsampling processing on the first feature and performing feature fusion processing on the fourth image feature to obtain a first fusion result; The first fusion result is upsampled and fused with the third image feature to obtain the second feature.

5. The method according to claim 1, characterized in that: The identifying the water leakage of the first pipeline based on the second result includes: Set the number of tokens in the preset token bucket to zero; determining the number of tokens based on the second result; In response to the number of tokens reaching a preset threshold, recording the condition of the first pipeline as a water leak; In response to the number of tokens being less than the preset threshold, the number of tokens is set to zero.

6. The method according to claim 5, characterized in that The determining the number of tokens based on the second result comprises: In response to the second result being a water leak, increasing the number of tokens by a first amount; In response to the second result being that there is no water leakage, the number of tokens is reduced by a second amount.

7. A water leakage identification device, characterized in that: include: A first acquisition module is configured to acquire an image at a first pipeline to obtain a first image; A first recognition module is configured to recognize the first image using a trained recognition model to obtain a first result; A second acquisition module is configured to, in response to the first result being water leakage, continue to acquire the image at the first pipeline to obtain at least one second image; A second recognition module is configured to recognize each of the second images using the trained recognition model to obtain at least one second result; The third identification module is configured to identify the water leakage of the first pipeline based on the second result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, comprising computer program instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 6.

Images