YOLOv10-based reservoir area floating object identification method, device and equipment, storage medium and product

Abstract

The invention provides a YOLOv10-based reservoir area floating object identification method, device and equipment, a storage medium and a product, relates to the technical field of environment monitoring, and aims to solve the problems that a traditional floating object identification method generally depends on manual observation, is low in efficiency and high in cost, and is difficult to meet the requirement of efficient management. The method comprises the steps that real-time detection is conducted on floaters according to a flotage recognition model, information of the floaters is obtained, and the information of the floaters comprises the position and the type; the floating object identification model is obtained by training a YOLOv10 model based on image data marked with floating object positions and types; the information of the floating objects is sent to a central control system for analysis and display, and the positions, types and the number of the floating objects are obtained; whether the number of the floating objects exceeds a preset threshold value or not is judged; and when the number of the floaters exceeds a preset threshold value, a corresponding alarm mechanism is triggered.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of environmental monitoring technology, and specifically, to a method, device, equipment, storage medium and product for identifying floating objects in a reservoir area based on YOLOv10. Background Art

[0002] Floating objects in reservoir areas play a vital role in the management of water bodies such as reservoirs and dams. They are not only related to water quality safety, but also directly affect the smoothness of water inlets and the stability of the overall ecological environment. Traditional floating object identification methods usually rely on manual observation, which is not only inefficient but also costly, and it is difficult to meet the needs of efficient management. Summary of the invention

[0003] The embodiments of the present application provide a method, device, equipment, storage medium and product for identifying floating objects in a reservoir area based on YOLOv10, aiming to solve the problem that traditional floating object identification methods usually rely on manual observation, which is not only inefficient and costly, but also difficult to meet the needs of efficient management.

[0004] The first aspect of the embodiment of the present application provides a method for identifying floating objects in a reservoir area based on YOLOv10, comprising:

[0005] Detecting floating objects in real time according to a floating object recognition model to obtain information about the floating objects, the information about the floating objects including: location and type; the floating object recognition model is obtained by training a YOLOv10 model based on image data marked with the location and type of the floating objects;

[0006] The information of the floating objects is sent to a central control system for analysis and display to obtain the location, type and quantity of the floating objects;

[0007] Determine whether the number of the floating objects exceeds a preset threshold; if the number of the floating objects exceeds the preset threshold, trigger a corresponding alarm mechanism.

[0008] In an optional embodiment, the floating object recognition model is trained by the following steps:

[0009] Acquiring image data of the floating object;

[0010] Annotating the image data, annotating the location and type of the floating object in the image data, and obtaining processed image data;

[0011] According to the processed image data, the YOLOv10 model is trained to obtain the trained floating object recognition model.

[0012] In an optional embodiment, after acquiring the image data of the floating object, the method further includes:

[0013] Performing data enhancement on the image data includes: randomly cropping, rotating, scaling and color transforming the image data.

[0014] In an optional implementation, the YOLOv10 model is trained according to the processed image data to obtain the trained floating object recognition model, including:

[0015] The processed image data is divided into a training set and a validation set. According to the training set and the validation set, the YOLOv10 model is repeatedly trained and validated to obtain the trained floating object recognition model.

[0016] In an optional embodiment, the real-time detection of floating objects according to the floating object recognition model to obtain the information of the floating objects includes:

[0017] Deploying the floating object recognition model onto an edge computing device;

[0018] The edge computing device collects real-time images of floating objects and inputs the real-time images into the floating object recognition model;

[0019] The floating object recognition model detects floating objects according to the real-time image to obtain information about the floating objects.

[0020] In an optional embodiment, the processed information is sent to a central control system for analysis and display to obtain the location, type and quantity of the floating objects, including:

[0021] The processed information is sent to a central control system, and the central control system analyzes the processed information and displays the data on a user interface; the data includes: the location, type, quantity and detection time of the floating objects.

[0022] A second aspect of an embodiment of the present application provides a reservoir floating object identification device based on YOLOv10, comprising:

[0023] A real-time detection module is used to detect floating objects in real time according to a floating object recognition model to obtain information about the floating objects, wherein the information about the floating objects includes: location and type; the floating object recognition model is obtained by training a YOLOv10 model based on image data marked with the location and type of the floating objects;

[0024] An analysis and display module is used to send the processed information to a central control system for analysis and display, and obtain the location, type and quantity of the floating objects;

[0025] The alarm module is used to determine whether the number of the floating objects exceeds a preset threshold; when the number of the floating objects exceeds the preset threshold, a corresponding alarm mechanism is triggered.

[0026] In an optional implementation, the real-time detection module is specifically used for:

[0027] Acquiring image data of the floating object;

[0028] Annotating the image data, annotating the location and type of the floating object in the image data, and obtaining processed image data;

[0029] According to the processed image data, the YOLOv10 model is trained to obtain the trained floating object recognition model.

[0030] In an optional implementation, the real-time detection module is specifically used for:

[0031] Performing data enhancement on the image data includes: randomly cropping, rotating, scaling and color transforming the image data.

[0032] In an optional implementation, the real-time detection module is specifically used for:

[0033] The processed image data is divided into a training set and a validation set. According to the training set and the validation set, the YOLOv10 model is repeatedly trained and validated to obtain the trained floating object recognition model.

[0034] In an optional implementation, the real-time detection module is specifically used for:

[0035] Deploying the floating object recognition model onto an edge computing device;

[0036] The edge computing device collects real-time images of floating objects and inputs the real-time images into the floating object recognition model;

[0037] The floating object recognition model detects floating objects according to the real-time image to obtain information about the floating objects.

[0038] In an optional implementation, the analysis and display module is specifically used for:

[0039] The processed information is sent to a central control system, and the central control system analyzes the processed information and displays the data on a user interface; the data includes: the location, type, quantity and detection time of the floating objects.

[0040] In a third aspect of an embodiment of the present application, an electronic device is provided, including: a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method for identifying floating objects in a reservoir area based on YOLOv10 according to the first aspect.

[0041] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for identifying floating objects in a reservoir area based on YOLOv10 described in the first aspect are implemented.

[0042] In a fifth aspect of the embodiments of the present application, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of the method for identifying floating objects in a reservoir area based on YOLOv10 as described in the first aspect.

[0043] In the disclosed embodiment, the floating objects are detected in real time by using the floating object recognition model based on YOLOv10, and the information of the floating objects is sent to the central control system for analysis and display, so as to obtain the location, type and quantity of the floating objects, and judge whether the number of floating objects exceeds the preset threshold. If the preset threshold is exceeded, the corresponding alarm mechanism is triggered. The floating objects in the reservoir area can be detected in real time and efficiently, and the full automation of floating object recognition is realized, which reduces the labor cost. The floating object recognition model can be deployed on a variety of edge computing devices to meet the monitoring needs of floating objects in the reservoir area under different environments. 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 description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0045] Figure 1 It is a flowchart of the steps of a method for identifying floating objects in a reservoir area based on YOLOv10 proposed in an embodiment of the present application;

[0046] Figure 2 It is a structural schematic diagram of a reservoir floating object identification device based on YOLOv10 proposed in an embodiment of the present application;

[0047] Figure 3 is a schematic diagram of an electronic device in the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] In the drawings, the size of the constituent elements, the thickness of the layer or the area may be exaggerated for the sake of clarity. Therefore, any implementation of the present disclosure is not necessarily limited to the size shown in the drawings, and the shapes and sizes of the components in the drawings do not reflect the true proportions. In addition, the drawings schematically show ideal examples, and any implementation of the present disclosure is not limited to the shapes or values ​​shown in the drawings.

[0050] Floating objects in reservoir areas are an important issue in the management of water bodies such as reservoirs and dams, which may affect water quality, block water inlets, and damage the ecological environment. Traditional floating object identification methods usually rely on manual observation, which is inefficient and costly.

[0051] The development of deep learning technology has provided new possibilities for the automatic identification of floating objects. As a real-time target detection method, the YOLO (You Only Look Once) series of algorithms have been gradually applied to various object detection tasks due to their high efficiency and accuracy. YOLOv10 is the latest version of the YOLO series, introducing a new real-time target detection method based on the YOLO series of algorithms.

[0052] Reference Figure 1 , Figure 1 1 is a flowchart of a method for identifying floating objects in a reservoir area based on YOLOv10 proposed in Example 1 of the present application. Figure 1 As shown, the method includes the following steps S11 to S13:

[0053] Step S11: Detect floating objects in real time according to the floating object recognition model to obtain information about the floating objects, wherein the information about the floating objects includes: location and type; the floating object recognition model is obtained by training the YOLOv10 model based on image data marked with the location and type of the floating objects.

[0054] YOLOv10 improves the model structure of previous versions of YOLO, and achieves efficient reasoning without NMS (non-maximum suppression) through a consistent dual allocation strategy and overall efficiency-accuracy driven model design. While maintaining high detection accuracy, it significantly improves detection speed, has higher efficiency and lower computational cost.

[0055] The model architecture of YOLOv10 consists of the following parts:

[0056] Backbone network: responsible for feature extraction, using an enhanced version of CSPNet (Cross-Stage Partial Network) to improve gradient flow and reduce computational redundancy. CSPNet splits the channels of feature maps and processes them through different paths, and finally merges the results, thereby enhancing the network's learning ability and reducing the amount of computation.

[0057] Neck: Aggregates features of different scales and passes them to the head. It includes the PAN (Path Aggregation Network) layer, which enables effective multi-scale feature fusion. The PAN layer passes low-level feature information to high-level features through bottom-up path enhancement, thereby improving the model's detection ability for objects of different scales.

[0058] Head: Generates the final detection results, including bounding boxes and category predictions. YOLOv10 adopts a dual label assignment strategy, namely One-to-Many Head and One-to-One Head. One-to-ManyHead: Generates multiple predictions for each object during training to provide rich supervision signals and improve learning accuracy. One-to-One Head: Generates an optimal prediction for each object during inference, eliminating the need for NMS (non-maximum suppression), reducing latency and improving efficiency.

[0059] Other key components: NMS-Free Training, which uses consistent dual assignment to eliminate the need for NMS, thereby reducing inference latency. Overall efficiency-accuracy driven model design, comprehensively optimizing various components from the perspective of efficiency and accuracy, including lightweight classification head, spatial channel decoupling down sampling, and level guidance block design. Enhanced model capabilities, incorporating large kernel convolution and partial self-attention modules to improve performance without adding a lot of computational cost. Large kernel convolution is used to enhance the feature extraction ability of the model, while partial self-attention modules are used to improve the global modeling ability of the model.

[0060] The floating object recognition model trained on the YOLOv10 model based on image data with the location and type of floating objects can detect floating objects in real time and output the information of floating objects. The information of floating objects includes: location (bounding box coordinates) and type (classification label). The floating object recognition model trained on YOLOv10 can improve the detection speed while maintaining high accuracy and achieve efficient real-time detection. This enables the model to process image data in real time in practical applications, which helps to quickly and accurately identify floating objects in images and provide strong support for subsequent decisions and actions.

[0061] Furthermore, the real-time detection of floating objects according to the floating object recognition model to obtain the information of the floating objects includes:

[0062] Deploying the floating object recognition model onto an edge computing device;

[0063] The edge computing device collects real-time images of floating objects and inputs the real-time images into the floating object recognition model;

[0064] The floating object recognition model detects floating objects according to the real-time image to obtain information about the floating objects.

[0065] The trained floating object recognition model is deployed on edge computing devices, such as drones, surveillance cameras, etc. The devices must have GPU (Graphics Processing Unit) acceleration capabilities to ensure the speed and accuracy of real-time detection. The edge computing device collects real-time images of floating objects. It can obtain real-time images by acquiring monitoring videos of the reservoir in real time and converting them into processable image frames. Then the real-time image is input into the floating object recognition model for floating object detection, and the model outputs the information of floating objects. The use of a variety of edge computing devices can adapt to the needs of reservoir floating object monitoring in different environments, which helps to improve the accuracy and robustness of floating object identification.

[0066] Furthermore, the floating object recognition model is trained by the following steps:

[0067] Acquiring image data of the floating object;

[0068] Annotating the image data, annotating the location and type of the floating object in the image data, and obtaining processed image data;

[0069] According to the processed image data, the YOLOv10 model is trained to obtain the trained floating object recognition model.

[0070] Collect image data containing floating objects in the reservoir area. The data sources can include video frames taken by surveillance cameras, images taken by drone cruises, etc. The collected floating object image data is the basis for machine learning model learning. Data from multiple sources are rich and diverse, thereby improving the generalization ability and recognition accuracy of the model.

[0071] To label the image data of floating objects, you can use labeling tools such as LabelImg, Yolo_mark, ImageLabel, etc. These tools support bounding box labeling of targets in images and can be saved as files in XML, JSON and other formats. By using the labeling tools, the location and type of floating objects are marked in the image data to obtain processed image data. The labeling process clearly marks the location and type of floating objects in the image data, providing a learning target for the model.

[0072] When the YOLOv10 model is trained, it will learn how to predict the location and type of floating objects based on the features of the processed input image data. By continuously iterating and optimizing the model parameters, the model's detection accuracy and speed are improved, and a trained floating object recognition model is obtained. The floating object recognition model trained based on YOLOv10 can achieve a faster detection speed while maintaining a high recognition accuracy, greatly improving the detection efficiency.

[0073] Furthermore, after obtaining the image data of the floating object, the method further includes:

[0074] Performing data enhancement on the image data includes: randomly cropping, rotating, scaling and color transforming the image data.

[0075] Random cropping: Randomly select an area in the image for cropping to generate a new image sample. The size and position of the crop can be randomly selected, but make sure that the cropped image still contains valid floating object information. You can use the cropping function provided by the image processing library (such as OpenCV, PIL, etc.) and combine it with a random number generator to determine the starting coordinates and size of the crop. Random cropping can increase the diversity of image data, enable the model to learn the characteristics of floating objects of different positions and sizes, and improve the robustness and generalization ability of the model.

[0076] Rotation: Rotate the image around its center point or a specified point at a random angle. The rotation angle can be any value, but it is usually randomly selected within a certain range (such as 0° to 360°). You can use the rotation function provided by the image processing library and combine it with a rotation matrix or affine transformation to achieve image rotation. Rotation can simulate the different angles that floating objects may appear in actual scenes, improve the model's robustness to rotation changes, increase the diversity of image data, and improve the model's generalization ability.

[0077] Scaling: Randomly scale the size of the image to generate image samples of different sizes. The scaling ratio can be randomly selected within a certain range, but it should be ensured that the scaled image does not distort or lose key information. You can use the scaling function provided by the image processing library and set the appropriate scaling ratio to achieve image scaling. Scaling enables the model to adapt to floating object images of different sizes, improves the scale invariance of the model, increases the diversity of image data, and helps improve the generalization ability of the model.

[0078] Color transformation: Randomly adjust the color attributes of the image, such as brightness, contrast, and saturation, to generate image samples after the color change. The degree of adjustment can be randomly selected within a certain range. The color transformation function provided by the image processing library can be used in combination with a random number generator to determine the degree of adjustment. Color transformation can simulate the impact of changes in lighting conditions on image color in actual scenes, improve the model's robustness to color changes, increase the diversity of image data, and help improve the model's recognition performance under different lighting conditions.

[0079] By performing data augmentation operations such as random cropping, rotation, scaling, and color transformation on image data, the diversity and quantity of image data can be significantly increased, thereby effectively improving the robustness, generalization ability, and recognition accuracy of the deep learning model (floating object recognition model).

[0080] Furthermore, the YOLOv10 model is trained according to the processed image data to obtain the trained floating object recognition model, including:

[0081] The processed image data is divided into a training set and a validation set. According to the training set and the validation set, the YOLOv10 model is repeatedly trained and validated to obtain the trained floating object recognition model.

[0082] The cross-validation method is used to adjust the model parameters, that is, to optimize the detection accuracy by dividing the processed image data into a training set and a validation set. Usually, the training set occupies the majority (such as 70% to 80%) and is used for model learning; the validation set occupies the remaining part (such as 20% to 30%) and is used to evaluate the performance of the model.

[0083] The YOLOv10 model is repeatedly trained and verified using the training set and the validation set to obtain a trained floating object recognition model. During the training process, the model will continuously adjust its internal parameters (such as weights and biases) to minimize the loss function (such as cross entropy loss or mean square error loss), thereby improving the recognition ability of floating objects. The training process requires multiple iterations (i.e., epochs), and each iteration will traverse the entire training set once. After each epoch or a certain number of iterations, the model is verified using the validation set. By calculating the performance indicators on the validation set (such as accuracy, recall rate, F1 score, average precision, etc.), the generalization ability and stability of the model can be understood. According to the performance indicators on the validation set, the model is tuned, including adjusting hyperparameters such as learning rate, batch size, anchor box, optimization algorithm, regularization term, and modifying the model structure. Dividing the processed image data into training set and validation set, and repeatedly training and verifying according to these data sets will help improve the generalization ability and recognition accuracy of the model, and also help to timely discover and solve problems, optimize model parameters, and improve model stability.

[0084] Step S12: Send the information of the floating objects to the central control system for analysis and display to obtain the location, type and quantity of the floating objects.

[0085] The processed information (including the location, type and quantity of floating objects) is packaged into a standard format and sent to the central control system via wired or wireless means (such as 4G / 5G network, Wi-Fi, satellite communication, etc.). The central control system receives the data (processed information) from each edge computing device and analyzes it, and visualizes the location, type and quantity of floating objects. The data visualization function provided by the central control system helps managers to intuitively understand the water conditions and provide a scientific basis for formulating cleanup plans and environmental protection policies.

[0086] Furthermore, the processed information is sent to a central control system for analysis and display to obtain the location, type and quantity of the floating objects, including:

[0087] The processed information is sent to a central control system, and the central control system analyzes the processed information and displays the data on a user interface; the data includes: the location, type, quantity and detection time of the floating objects.

[0088] The processed information is sent to the central control system via wired or wireless means. The central control system receives the information from each edge computing device for analysis, and visualizes the location, type, quantity, detection time and other information of floating objects on the user interface. It can also update the distribution of floating objects in the reservoir area in real time.

[0089] Specifically, the steps of sending the processed floating object information to the central control system and updating it in real time involve several key technical links, such as data transmission, interface design, information display and real-time update. The following are the specific implementation methods:

[0090] Data transmission: Select the appropriate network transmission protocol. According to the real-time requirements of the system and the network environment, select the most suitable protocol. For example, WebSocket can be used to update data in real time, while MQTT is suitable for low-bandwidth, high-latency network environments. Data formatting: Use JSON (JavaScript Object Notation) format to organize information such as the type, location (such as latitude and longitude or relative coordinates) and timestamp of floating objects. JSON format is easy to parse and use across platforms, and is suitable for Web (WorldWide Web, global wide area network) applications.

[0091] Interface design: Design a RESTful (Representational State Transfer) style API (Application Programming Interface) to allow the central control system to query and receive data through standard HTTP methods (GET, POST, PUT, DELETE, etc.). If the WebSocket protocol is used, design a message push mechanism so that when the recognition system detects a new floating object, it can actively push data to the central control system. Security design uses SSL / TLS to encrypt data transmission to ensure the security of data during transmission. Design a suitable authentication mechanism (such as OAuth2.0) to ensure that only authorized users or systems can access the API.

[0092] Information display: User interface design, integrate maps (such as Google Maps, OpenStreetMap) in the user interface of the central control system, use the map to display the location of floating objects, and design the interface to dynamically display the latest data, including the type, location, quantity and detection time of floating objects. For example, there are various floating object marks on the map of the user interface. When a mark is selected, the location on the map will be displayed. Figure 1 The window can be used as a pop-up window or side window to display detailed information of the currently selected mark, including the type of floating objects, location coordinates, quantity and detection time, etc. Interactive function provides filtering and query functions, allowing users to view historical and real-time data according to time, type and other conditions.

[0093] Real-time update: Database design, use a database that supports real-time updates (such as Firebase RealtimeDatabase, MongoDB). When new data is written, the database automatically pushes updates to all connected clients. Design data caching strategies to reduce the number of database queries and improve system response speed. Back-end architecture, the back-end adopts an event-driven architecture to ensure that when new data arrives, it can be quickly processed and updated to the user interface.

[0094] Through the above steps, data transmission, analysis and display, and real-time updating from floating object identification to the central control system can be efficiently realized, ensuring the timeliness and accuracy of reservoir area management.

[0095] Step S13: determining whether the number of the floating objects exceeds a preset threshold; if the number of the floating objects exceeds the preset threshold, triggering a corresponding alarm mechanism.

[0096] The central control system makes a judgment based on the preset floating object number threshold (set according to water area type, environmental protection requirements and other factors). If the number of floating objects detected exceeds the preset threshold, the alarm mechanism is triggered. The alarm mechanism includes alarm mode and emergency response. Alarm mode: timely send alarm information to relevant managers or emergency response teams through sound and light alarm devices, SMS notifications, email reminders or APP push, etc. Emergency response: managers quickly formulate and implement cleaning plans based on alarm information, and mobilize cleaning equipment and personnel to go to the scene for processing when necessary. By setting the floating object number threshold and triggering the alarm mechanism, it can issue a warning in time when the number of floating objects exceeds the standard, and clean up floating objects in the water area in time, which will help reduce water pollution, protect the health of aquatic ecosystems, and promote sustainable development.

[0097] For ease of understanding, the embodiments of the present disclosure are described through specific application scenarios.

[0098] Edge computing devices such as surveillance cameras and drones are deployed at key locations such as the reservoir inlet, outlet, central area, and shores where floating objects are easily accumulated. The edge computing devices on the cameras and drones continuously capture the reservoir images, and the built-in graphics processor processes the video stream in real time to extract real-time images containing floating objects. The real-time images are then input into the floating object recognition model deployed in the device for floating object detection, and the type (such as plastic bottles, branches, algae, etc.) and location of the floating objects are output, and the information is sent to the central control system in real time through the data transmission network. The information data received by the central control system is analyzed and processed on the server, and then the location of the floating objects is displayed as a mark on the integrated map of the user interface. When the user selects a mark, the detailed information of the currently selected mark is displayed on one side of the map area, including the type, location coordinates, quantity, and detection time of floating object A. The system calculates the number of all current floating objects regularly or in real time. If it exceeds the preset threshold (such as no more than 100 floating objects per square kilometer), the sound and light alarm device is immediately triggered, and an emergency notification window pops up through the user interface to remind the management personnel to take immediate measures to clean up the floating objects in the reservoir. After the cleaning is completed, the management personnel will enter the cleaning results into the system. The system will automatically adjust the parameters of the floating object identification model or optimize the monitoring strategy based on the feedback data to improve the accuracy and efficiency of future monitoring.

[0099] Based on the same inventive concept, the present application embodiment discloses a reservoir floating object recognition device based on YOLOv10, such as Figure 2 As shown, the device comprises:

[0100] A real-time detection module is used to detect floating objects in real time according to a floating object recognition model to obtain information about the floating objects, wherein the information about the floating objects includes: location and type; the floating object recognition model is obtained by training a YOLOv10 model based on image data marked with the location and type of the floating objects;

[0101] An analysis and display module is used to send the information of the floating objects to a central control system for analysis and display, and obtain the location, type and quantity of the floating objects;

[0102] The alarm module is used to determine whether the number of the floating objects exceeds a preset threshold; when the number of the floating objects exceeds the preset threshold, a corresponding alarm mechanism is triggered.

[0103] In an optional implementation, the real-time detection module is specifically used for:

[0104] Acquiring image data of the floating object;

[0105] Annotating the image data, annotating the location and type of the floating object in the image data, and obtaining processed image data;

[0106] According to the processed image data, the YOLOv10 model is trained to obtain the trained floating object recognition model.

[0107] In an optional implementation, the real-time detection module is specifically used for:

[0108] Performing data enhancement on the image data includes: randomly cropping, rotating, scaling and color transforming the image data.

[0109] In an optional implementation, the real-time detection module is specifically used for:

[0110] The processed image data is divided into a training set and a validation set. According to the training set and the validation set, the YOLOv10 model is repeatedly trained and validated to obtain the trained floating object recognition model.

[0111] In an optional implementation, the real-time detection module is specifically used for:

[0112] Deploying the floating object recognition model onto an edge computing device;

[0113] The edge computing device collects real-time images of floating objects and inputs the real-time images into the floating object recognition model;

[0114] The floating object recognition model detects floating objects according to the real-time image to obtain information about the floating objects.

[0115] In an optional implementation, the analysis and display module is specifically used for:

[0116] The processed information is sent to a central control system, and the central control system analyzes the processed information and displays the data on a user interface; the data includes: the location, type, quantity and detection time of the floating objects.

[0117] The YOLOv10-based reservoir area floating object identification method provided by the disclosed embodiment detects floating objects in real time through a YOLOv10-based floating object identification model, obtains floating object information and sends it to a central control system for analysis and display, obtains the location, type and quantity of floating objects, and determines whether the number of floating objects exceeds a preset threshold. If it exceeds the preset threshold, a corresponding alarm mechanism is triggered. It can detect floating objects in the reservoir area in real time and efficiently, realizes full automation of floating object identification, reduces labor costs, and the floating object identification model can be deployed on a variety of edge computing devices to meet the needs of reservoir area floating object monitoring in different environments.

[0118] It should be noted that, for the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present disclosure are not limited by the order of the actions described, because according to the embodiments of the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present disclosure.

[0119] The present disclosure also provides an electronic device, referring to Figure 3 , Figure 3 Schematic diagram of an electronic device shown in an embodiment of the present disclosure. Figure 3 As shown, the electronic device 100 includes: a memory 110 and a processor 120. The memory 110 and the processor 120 are connected via a bus communication. A computer program is stored in the memory 110. The computer program can be run on the processor 120 to implement the steps in the method for identifying floating objects in a reservoir area based on YOLOv10 disclosed in an embodiment of the present disclosure.

[0120] The embodiment of the present disclosure also provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of a computer device, the computer device is enabled to perform the steps in the method for identifying floating objects in a reservoir area based on YOLOv10 as disclosed in the embodiment of the present disclosure.

[0121] The embodiment of the present disclosure also provides a computer program product, including a computer program. When the computer program is executed by a processor of a computer device, the computer program can execute the steps in the method for identifying floating objects in a reservoir area based on YOLOv10 as disclosed in the embodiment of the present disclosure.

[0122] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0123] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, apparatuses, electronic devices, and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0126] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0127] Finally, 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 terminal 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 terminal device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0128] The above is a detailed introduction to a method, device, equipment, storage medium and product for identifying floating objects in a reservoir area based on YOLOv10 provided by the present application. In this article, specific examples are used to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application; at the same time, for those skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for identifying floating objects in a reservoir area based on YOLOv10, characterized in that: include: Detecting floating objects in real time according to a floating object recognition model to obtain information about the floating objects, the information about the floating objects including: location and type; the floating object recognition model is obtained by training a YOLOv10 model based on image data marked with the location and type of the floating objects; The information of the floating objects is sent to a central control system for analysis and display to obtain the location, type and quantity of the floating objects; Determine whether the number of the floating objects exceeds a preset threshold; if the number of the floating objects exceeds the preset threshold, trigger a corresponding alarm mechanism.

2. The method for identifying floating objects in a reservoir area based on YOLOv10 according to claim 1, characterized in that: The floating object recognition model is trained by following steps: Acquiring image data of the floating object; Annotating the image data, annotating the location and type of the floating object in the image data, and obtaining processed image data; According to the processed image data, the YOLOv10 model is trained to obtain the trained floating object recognition model.

3. The method for identifying floating objects in a reservoir area based on YOLOv10 according to claim 2 is characterized in that: After obtaining the image data of the floating object, the method further includes: Performing data enhancement on the image data includes: randomly cropping, rotating, scaling and color transforming the image data.

4. The method for identifying floating objects in a reservoir area based on YOLOv10 according to claim 2, characterized in that: The YOLOv10 model is trained according to the processed image data to obtain the trained floating object recognition model, including: The processed image data is divided into a training set and a validation set. According to the training set and the validation set, the YOLOv10 model is repeatedly trained and validated to obtain the trained floating object recognition model.

5. The method for identifying floating objects in a reservoir area based on YOLOv10 according to claim 1, characterized in that: The real-time detection of floating objects according to the floating object recognition model to obtain the information of the floating objects includes: Deploying the floating object recognition model onto an edge computing device; The edge computing device collects real-time images of floating objects and inputs the real-time images into the floating object recognition model; The floating object recognition model detects floating objects according to the real-time image to obtain information about the floating objects.

6. The method for identifying floating objects in a reservoir area based on YOLOv10 according to claim 1, characterized in that: The processed information is sent to a central control system for analysis and display to obtain the location, type and quantity of the floating objects, including: The processed information is sent to a central control system, and the central control system analyzes the processed information and displays the data on a user interface; the data includes: the location, type, quantity and detection time of the floating objects.

7. A floating object recognition device in a reservoir area based on YOLOv10, characterized in that: The device comprises: A real-time detection module is used to detect floating objects in real time according to a floating object recognition model to obtain information about the floating objects, wherein the information about the floating objects includes: location and type; the floating object recognition model is obtained by training a YOLOv10 model based on image data marked with the location and type of the floating objects; An analysis and display module is used to send the information of the floating objects to a central control system for analysis and display, and obtain the location, type and quantity of the floating objects; The alarm module is used to determine whether the number of the floating objects exceeds a preset threshold; when the number of the floating objects exceeds the preset threshold, a corresponding alarm mechanism is triggered.

8. The reservoir floating object identification device based on YOLOv10 according to claim 7 is characterized in that: The real-time detection module is specifically used for: Acquiring image data of the floating object; Annotating the image data, annotating the location and type of the floating object in the image data, and obtaining processed image data; The YOLOv10 model is trained according to the processed image data to obtain the trained floating object recognition model.

9. The reservoir floating object identification device based on YOLOv10 according to claim 8 is characterized in that: The real-time detection module is specifically used for: Performing data enhancement on the image data includes: randomly cropping, rotating, scaling and color transforming the image data.

10. The reservoir floating object identification device based on YOLOv10 according to claim 8, characterized in that: The real-time detection module is specifically used for: The processed image data is divided into a training set and a validation set. According to the training set and the validation set, the YOLOv10 model is repeatedly trained and validated to obtain the trained floating object recognition model.

11. The reservoir floating object identification device based on YOLOv10 according to claim 7, characterized in that: The real-time detection module is specifically used for: Deploying the floating object recognition model onto an edge computing device; The edge computing device collects real-time images of floating objects and inputs the real-time images into the floating object recognition model; The floating object recognition model detects floating objects according to the real-time image to obtain information about the floating objects.

12. The reservoir floating object identification device based on YOLOv10 according to claim 7, characterized in that: The analysis and display module is specifically used for: The processed information is sent to a central control system, and the central control system analyzes the processed information and displays the data on a user interface; the data includes: the location, type, quantity and detection time of the floating objects.

13. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for identifying floating objects in a reservoir area based on YOLOv10 according to any one of claims 1 to 6 is implemented.

14. A computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for identifying floating objects in a reservoir area based on YOLOv10 as described in any one of claims 1 to 6.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for identifying floating objects in a reservoir area based on YOLOv10 as described in any one of claims 1 to 6 is implemented.

Images