Oilfield produced water transportation monitoring video detection method and related device
Through the improved YOLOv8 and SlowFast models combined with text information classification module, the target detection and behavior analysis of oilfield water transport videos is realized, and the problem of low detection efficiency of oilfield water transport monitoring videos in the existing technology is solved, and the detection efficiency is improved and accuracy reduce labor costs, promote the safe and stable operation of oil fields and environmental protection.
Patent Information
- Application Number
- CN202411066024.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-22
AI Technical Summary
The existing video surveillance technology based on target detection and behavioral analysis lacks time series behavior analysis capabilities in oil field production water transport, and has low processing efficiency, so it is unable to effectively monitor the standardization of oil field production water transport operations, resulting in high labor costs and economical Low efficiency.
Using the improved YOLOv8 object detection model and SlowFast behavior recognition model, the operation specification detection and recognition model is constructed by adding attention mechanisms and replacing the loss function in the backbone network as the focus loss function, and combining the text information classification module to build an operation specification detection and recognition model to achieve the extraction of water from the oil field. Target detection, behavioral analysis and process standardization detection of videos.
It improves the efficiency and accuracy of the monitoring video inspection of oilfield water extraction and transportation, and can dynamically detect behavior normatively, reduce calculation amount, reduce labor costs, and promote the safe and stable operation of oilfields. It has important environmental protection significance.
Smart Images

Figure CN119007287A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of image processing and relates to an oilfield produced water transportation monitoring video detection method and a related device. Background Art
[0002] Oil fields produce a large amount of produced water every day. Produced water is a mixture of oil and water. Direct discharge will cause serious pollution to the environment. Therefore, oil fields will store produced water and regularly transport it to sewage treatment plants for sewage discharge and purification. Therefore, oil fields need to transport a large amount of produced water regularly. Whether the produced water transportation operation can be standardized is crucial to the environment and safe production. The staff uses the full-coverage video surveillance installed in the station to supervise whether the station employees are transporting oil field produced water in a standardized manner, including status detection, behavior detection and process detection.
[0003] Manual monitoring requires a lot of manpower and has low economic benefits. With the emergence of deep learning technologies such as target detection, behavior analysis, and text classification, manual monitoring is gradually being replaced. The existing video behavior detection technology based on target detection has the ability to analyze a certain instantaneous picture in the surveillance video and predict whether the behavior is normal by analyzing the instantaneous picture. This technology does not have the ability to analyze behavior in time series. However, the existing video behavior detection technology based on behavior analysis has low processing efficiency, fails to analyze the behavior of key targets in the video, and does not have the ability to analyze the logical relationship between different behaviors. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method and related device for monitoring video detection of oilfield produced water transportation.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] In a first aspect, the present invention provides a method for detecting a monitoring video of oilfield produced water transportation, comprising: obtaining a monitoring video of oilfield produced water transportation; inputting the monitoring video of oilfield produced water transportation into a pre-trained operation specification detection and recognition model to obtain a standard detection result of oilfield produced water transportation; wherein the operation specification detection and recognition model comprises a target detection module, a behavior analysis module and a text information classification module connected in sequence; the target detection module is constructed based on an improved YOLOv8 target detection model, and the improved YOLOv8 target detection model is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model; the behavior analysis module is constructed based on an improved SlowFast behavior recognition model, and the improved SlowFast behavior recognition model is constructed by replacing the loss function of the SlowFast behavior recognition model with a focal loss function Focal Loss is obtained; the target detection module is used to obtain the information of each target to be detected and the analysis scope of each behavior to be detected in the oilfield produced water transportation monitoring video, and to obtain the state normative information according to the information of each target to be detected; the behavior analysis module is used to obtain the information of each behavior to be detected according to the information of each target to be detected and the analysis scope of each behavior to be detected, and to obtain the behavior normative information according to the information of each behavior to be detected; the text information classification module is used to connect the information of each target to be detected and the information of each behavior to be detected into text and classify it to obtain the process normative information.
[0007] Optionally, it also includes: adding a small target detection head to the improved YOLOv8 target detection model.
[0008] Optionally, the focal loss function is:
[0009] FL(P t )=-a t (1-P t ) γ log(P t )
[0010] Among them, FL(P t ) is the loss value of the focal loss function Focal Loss, P t To improve the prediction probability of the SlowFast behavior recognition model, a t is the preset weight factor, and γ is a parameter greater than 0 and less than 5.
[0011] Optionally, the text information classification module is constructed based on an LSTM model.
[0012] Optionally, the obtaining of state standardization according to each target information to be detected includes: when there is a target with irregular state in each target information to be detected, the state standardization is irregular state and a irregular state alarm signal is generated; wherein, the targets with irregular state include persons without helmets, persons without protective clothing and persons smoking.
[0013] Optionally, the obtaining of behavior normative information based on each behavior information to be detected includes: when there is irregular behavior in each behavior information to be detected, the behavior normative information is irregular behavior and a irregular behavior alarm signal is generated; wherein, the irregular behavior includes produced water leakage and personnel falling.
[0014] Optionally, it also includes: visually displaying the standardized inspection results of oilfield produced water transportation in the form of a web interface and / or a desktop application interface.
[0015] In a second aspect, the present invention provides an oilfield produced water transportation monitoring video detection system, including: a data acquisition module, used to acquire oilfield produced water transportation monitoring video; a specification detection module, used to input the oilfield produced water transportation monitoring video into a pre-trained operation specification detection and recognition model to obtain the oilfield produced water transportation standard detection result; wherein the operation specification detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module connected in sequence; the target detection module is constructed based on an improved YOLOv8 target detection model, and the improved YOLOv8 target detection model is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model; the behavior analysis module is constructed based on an improved SlowFast behavior recognition model, and the improved SlowFast behavior recognition model is constructed by replacing the loss function of the SlowFast behavior recognition model with a focal loss function Focal Loss is obtained; the target detection module is used to obtain the information of each target to be detected and the analysis scope of each behavior to be detected in the oilfield produced water transportation monitoring video, and to obtain the state normative information according to the information of each target to be detected; the behavior analysis module is used to obtain the information of each behavior to be detected according to the information of each target to be detected and the analysis scope of each behavior to be detected, and to obtain the behavior normative information according to the information of each behavior to be detected; the text information classification module is used to connect the information of each target to be detected and the information of each behavior to be detected into text and classify it to obtain the process normative information.
[0016] A third aspect of the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned oilfield produced water transportation monitoring video detection method when executing the computer program.
[0017] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned oilfield produced water transportation monitoring video detection method are implemented.
[0018] Compared with the prior art, the present invention has the following beneficial effects:
[0019] The oilfield produced water transportation monitoring video detection method of the present invention is based on the oilfield produced water transportation monitoring video, and realizes the standardized detection of oilfield produced water transportation through a pre-trained operation standard detection and recognition model. The operation standard detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module connected in sequence, which realizes the standardized detection capability of a certain state and behavior of the operation in the operation, and also has the analysis capability of whether the complete operation operation process is standardized. First, the target detection module obtains the analysis range of each target information to be detected and each behavior to be detected in the oilfield produced water transportation monitoring video, and obtains the state standardization information according to each target information to be detected, which can not only detect whether the static state information in the produced water transportation operation meets the requirements, but also reduce the calculation amount of behavior analysis and increase the operation speed. Then, the behavior analysis module obtains each behavior information to be detected according to the analysis range of each target information to be detected and each behavior to be detected, and obtains the behavior standardization information according to each behavior information to be detected, so as to perform dynamic behavior detection. Finally, the text information classification module connects the target information and behavior information to be detected into text to classify whether the operation is standardized or not, and obtains process standardization information. This can greatly promote the safe and stable operation of oil fields, reduce labor costs, and is also of great significance to environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 This is a flow chart of a method for monitoring video detection of oilfield produced water transportation according to an embodiment of the present invention.
[0021] Figure 2 It is a schematic diagram of the process of loading, transporting and unloading produced water from an oil field according to an embodiment of the present invention.
[0022] Figure 3 It is a schematic diagram of the main contents of the oilfield produced water transportation monitoring video according to an embodiment of the present invention.
[0023] Figure 4 It is a structural block diagram of the oilfield produced water transportation monitoring video detection system according to an embodiment of the present invention. DETAILED DESCRIPTION
[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0026] The present invention is further described in detail below in conjunction with the accompanying drawings:
[0027] See also Figure 1 In one embodiment of the present invention, a method for monitoring video detection of oilfield produced water transportation is provided, comprising the following steps:
[0028] S1: Obtain monitoring video of oilfield produced water transportation.
[0029] S2: Input the monitoring video of oilfield produced water transportation into the pre-trained operation specification detection and recognition model to obtain the standard detection result of oilfield produced water transportation.
[0030] Among them, the operation specification detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module which are connected in sequence; the target detection module is constructed based on the improved YOLOv8 target detection model, which is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model; the behavior analysis module is constructed based on the improved SlowFast behavior recognition model, which is obtained by replacing the loss function of the SlowFast behavior recognition model with the focal loss function Focal Loss; the target detection module is used to obtain the information of each target to be detected and the analysis scope of each behavior to be detected in the oilfield produced water transportation monitoring video, and to obtain state normative information according to the information of each target to be detected; the behavior analysis module is used to obtain the information of each behavior to be detected according to the information of each target to be detected and the analysis scope of each behavior to be detected, and to obtain behavior normative information according to the information of each behavior to be detected; the text information classification module is used to connect the information of each target to be detected and the information of each behavior to be detected into text and classify it to obtain process normative information.
[0031] Specifically, the improved YOLOv8 target detection model is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model. The attention mechanism is an important technology in deep learning, which can help the model better focus on the key information of the data, thereby improving the performance of the model.
[0032] The improved SlowFast behavior recognition model is obtained by replacing the loss function of the SlowFast behavior recognition model with the focal loss function. The SlowFast behavior recognition model uses the balanced cross entropy (BCE) loss function. Although BCE solves the imbalance problem of positive and negative samples, it does not distinguish between simple and difficult samples. When there are a large number of easily distinguishable negative samples, the entire training process will revolve around the easily distinguishable negative samples, which will then overwhelm the positive samples and cause a large loss.
[0033] The application scenario of the present invention has the problem of unbalanced data sets, and there are more easily distinguishable samples and fewer difficult-to-distinguish samples. In order to avoid the training process focusing on a large number of easily distinguishable samples and avoid causing large losses, the present invention improves the loss function of the SlowFast behavior recognition model and uses the Focal Loss loss function. Focal Loss introduces a modulation factor to focus on difficult-to-distinguish samples to achieve better behavior analysis capabilities.
[0034] The oilfield produced water transportation monitoring video detection method of the present invention is based on the oilfield produced water transportation monitoring video, and realizes the standardized detection of oilfield produced water transportation through a pre-trained operation standard detection and recognition model. The operation standard detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module connected in sequence, which realizes the standardized detection capability of a certain state and behavior of the operation in the operation, and also has the analysis capability of whether the complete operation operation process is standardized. First, the target detection module obtains the analysis range of each target information to be detected and each behavior to be detected in the oilfield produced water transportation monitoring video, and obtains the state standardization information according to each target information to be detected, which can not only detect whether the static state information in the produced water transportation operation meets the requirements, but also reduce the calculation amount of behavior analysis and increase the operation speed. Then, the behavior analysis module obtains each behavior information to be detected according to the analysis range of each target information to be detected and each behavior to be detected, and obtains the behavior standardization information according to each behavior information to be detected, so as to perform dynamic behavior detection. Finally, the text information classification module connects the target information and behavior information to be detected into text to classify whether the operation is standardized or not, and obtains process standardization information. This can greatly promote the safe and stable operation of oil fields, reduce labor costs, and is also of great significance to environmental protection.
[0035] In a possible implementation, the method further includes adding a small target detection head to the improved YOLOv8 target detection model.
[0036] Specifically, a small target detection head is added to the improved YOLOv8 target detection model to enhance the feature recognition and analysis of small targets in monitoring. YOLOv8 may contain multiple detection heads (such as P3, P4, and P5, etc.) by default, corresponding to feature maps of different scales for detecting targets of different sizes. In order to enhance the detection capability of small targets, a new detection head, such as P2, can be added, whose corresponding feature map has a higher resolution, so that more feature information of small targets can be captured.
[0037] In a possible implementation, the focal loss function is:
[0038] FL(P t )=-a t (1-P t ) γ log(P t )
[0039] Among them, FL(P t ) is the loss value of the focal loss function Focal Loss, P t To improve the prediction probability of the SlowFast behavior recognition model, a tis the preset weight factor, and γ is a parameter greater than 0 and less than 5.
[0040] Among them, a t As a preset weight factor, it is used to suppress the imbalance in the number of positive and negative samples.
[0041] In a possible implementation, the text information classification module is constructed based on an LSTM model.
[0042] Specifically, the behavior logic classification based on the LSTM model can be used for the detection of the standardization of oilfield produced water transportation, and can also analyze the temporal and logical relationships between behaviors.
[0043] Specifically, when training the text information classification module, the text data set it relies on includes two types of texts: one is the text formed by connecting the target information to be detected and the behavior information to be detected generated by the target detection module and the behavior analysis module from the monitoring video of oil field produced water transportation obtained in the produced water transportation operation under the standard process; the other is the text formed by connecting the target information to be detected and the behavior information to be detected generated by the target detection module and the behavior analysis module from the monitoring video of oil field produced water transportation obtained in the produced water transportation operation under the non-standard process. The text information classification module is trained based on the two types of text pairs, so that the text information classification module has the ability to analyze and predict whether the process is standardized.
[0044] In one possible implementation, see Figure 2 , showing the process of loading, transporting and unloading produced water from the oil field. Figure 3 , showing the main contents of the oilfield produced water transportation monitoring video, including state detection, behavior detection and process detection. Among them, the targets to be detected include personnel and equipment; the behaviors to be detected include: checking the environment and equipment, transferring and unloading produced water, connecting and disconnecting the produced water transportation pipeline, produced water leakage, and personnel falling.
[0045] In a possible implementation, the acquisition of state standardization based on the information of each target to be detected includes: when there is a target with irregular state in the information of each target to be detected, the state standardization is irregular state and a irregular state alarm signal is generated; wherein, the targets to be detected include personnel and equipment; targets with irregular state include personnel without helmets, personnel without protective clothing, and personnel smoking.
[0046] Specifically, when training the target detection module, the image data set it relies on can be made from the oilfield produced water transportation monitoring images transmitted back to the monitoring center. The set of photos of people wearing helmets and not wearing helmets, the set of photos of people wearing protective clothing and not wearing protective clothing, and the set of photos of people smoking can be manually found and labeled, and then used for the training of the target detection module. Optionally, an automatic labeling model can also be used for automatic labeling to effectively reduce labor costs.
[0047] In a possible implementation, the obtaining of behavior normative information based on each behavior information to be detected includes: when there is irregular behavior in each behavior information to be detected, the behavior normative information is irregular behavior and a irregular behavior alarm signal is generated; wherein, the irregular behavior includes produced water leakage and personnel falling.
[0048] Specifically, when training the behavior analysis module, the behavior data set it relies on can be made from the oilfield produced water transportation monitoring video transmitted back to the monitoring center. The behaviors to be detected in the oilfield produced water transportation monitoring video can be manually found and edited and annotated as a behavior data set. Optionally, the behavior recognition model is also used for preliminary annotation, which is then manually reviewed and adjusted.
[0049] In a possible implementation, the oilfield produced water transportation monitoring video detection method further includes: visually displaying the standardized detection results of the oilfield produced water transportation in the form of a web interface and / or a desktop application interface.
[0050] Specifically, visualization is performed in the form of a web interface and / or desktop application interface to meet the needs of different business scenarios. The web interface uses modern web site development technology, including front-end technology (HTML, CSS, JavaScript, etc.) and back-end technology (Spring, Springboot, Mybatis, etc.) to ensure the user-friendliness and responsiveness of the web interface. The desktop application interface can be designed and developed based on PyQt, providing a good user experience and convenient operation.
[0051] In a possible implementation, the integration and construction of the operation specification detection and recognition model includes:
[0052] Use the PyTorch deep learning framework to build, train, and deploy deep learning models. Use the torchvision library to implement visual functions. The library provides implementations of various classic visual models (including YOLO), as well as common data conversion and data loading functions. Use NumPy to process data in neural networks. NumPy is a basic package for scientific computing in Python. It provides support for multidimensional arrays and matrices, which is suitable for processing model input and output data. Use the opencv-python library for computer vision processing. OpenCV is an open source library for computer vision tasks. It provides various image processing and video processing functions that can help you process and preprocess video data. Use matplotlib to visualize the training effect. This library can be used to draw charts and visualize data. Common libraries help analyze the performance and results of the model. Use scikit-learn model evaluation and indicator calculation. scikit-learn is a very useful machine learning library that provides a variety of common evaluation indicators and model selection tools. Use tqdm to supervise the training process. tqdm is a library for displaying progress bars that can help you monitor the progress of model training and data processing. Through the above framework and deep learning package, the model of the present invention is built using Python language.
[0053] In specific applications, the trained operation specification detection and recognition model can be integrated into the monitoring system to form a serial structure detection model. The system is deployed at the oil field produced water transportation operation site to conduct testing and tuning in the actual environment. The system is optimized according to the test results to ensure the stability and reliability of the system. The data set is updated regularly and the operation specification detection and recognition model is retrained to improve the system recognition accuracy.
[0054] The present invention can be applied to auxiliary monitoring of oil fields, helping the central monitoring room to intelligently monitor and identify whether the produced water transportation operation in the oil field is standardized, and promptly alert the staff when irregular behavior is identified. The present invention can greatly promote the safe and stable operation of oil fields, reduce labor costs, and is also of great significance to environmental protection. The method is applied to auxiliary monitoring of oil fields, enabling oil fields to gradually transform into intelligent and digital ones.
[0055] The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.
[0056] See also Figure 4 In another embodiment of the present invention, a system for monitoring video detection of oilfield produced water transportation is provided, which can be used to implement the above-mentioned method for monitoring video detection of oilfield produced water transportation. Specifically, the system for monitoring video detection of oilfield produced water transportation includes a data acquisition module and a specification detection module.
[0057] Among them, the data acquisition module is used to obtain the monitoring video of oilfield produced water transportation; the standard detection module is used to input the monitoring video of oilfield produced water transportation into the pre-trained operation standard detection and recognition model to obtain the standard detection results of oilfield produced water transportation. Among them, the operation specification detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module which are connected in sequence; the target detection module is constructed based on the improved YOLOv8 target detection model, which is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model; the behavior analysis module is constructed based on the improved SlowFast behavior recognition model, which is obtained by replacing the loss function of the SlowFast behavior recognition model with the focal loss function Focal Loss; the target detection module is used to obtain the information of each target to be detected and the analysis scope of each behavior to be detected in the oilfield produced water transportation monitoring video, and to obtain state normative information according to the information of each target to be detected; the behavior analysis module is used to obtain the information of each behavior to be detected according to the information of each target to be detected and the analysis scope of each behavior to be detected, and to obtain behavior normative information according to the information of each behavior to be detected; the text information classification module is used to connect the information of each target to be detected and the information of each behavior to be detected into text and classify it to obtain process normative information.
[0058] All relevant contents of each step involved in the aforementioned embodiment of the oilfield produced water transportation monitoring video detection method can be referred to the functional description of the functional module corresponding to the oilfield produced water transportation monitoring video detection system in the embodiment of the present invention, and will not be repeated here.
[0059] The division of modules in the embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.
[0060] In another embodiment of the present invention, a computer device is provided, the computer device including a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the oilfield produced water transportation monitoring video detection method.
[0061] In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the oilfield produced water transportation monitoring video detection method in the above embodiment.
[0062] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0063] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of 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 device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing 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 1 A device that provides the functions specified in a block or multiple blocks.
[0064] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising 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.
[0065] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing 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.
[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for monitoring video detection of oilfield produced water transportation, characterized in that: include: Obtain monitoring video of oilfield produced water transportation; The monitoring video of oilfield produced water transportation is input into the pre-trained operation standard detection and recognition model to obtain the standard detection results of oilfield produced water transportation; Among them, the operation specification detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module which are connected in sequence; the target detection module is constructed based on the improved YOLOv8 target detection model, which is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model; the behavior analysis module is constructed based on the improved SlowFast behavior recognition model, which is obtained by replacing the loss function of the SlowFast behavior recognition model with the focal loss function Focal Loss; the target detection module is used to obtain the information of each target to be detected and the analysis scope of each behavior to be detected in the oilfield produced water transportation monitoring video, and to obtain state normative information according to the information of each target to be detected; the behavior analysis module is used to obtain the information of each behavior to be detected according to the information of each target to be detected and the analysis scope of each behavior to be detected, and to obtain behavior normative information according to the information of each behavior to be detected; the text information classification module is used to connect the information of each target to be detected and the information of each behavior to be detected into text and classify it to obtain process normative information.
2. The oilfield produced water transportation monitoring video detection method according to claim 1 is characterized in that: Also includes: Add a small target detection head to the improved YOLOv8 target detection model.
3. The oilfield produced water transportation monitoring video detection method according to claim 1, characterized in that: The focal loss function Focal Loss is: FL(P t )=-a t (1-P t ) γ log(P t ) Among them, FL(P t ) is the loss value of the focal loss function Focal Loss, P t To improve the prediction probability of the SlowFast behavior recognition model, a t is the preset weight factor, and γ is a parameter greater than 0 and less than 5.
4. The oilfield produced water transportation monitoring video detection method according to claim 1, characterized in that: The text information classification module is built based on the LSTM model.
5. The oilfield produced water transportation monitoring video detection method according to claim 1, characterized in that: The state standardization of obtaining the state standardization according to each target information to be detected includes: When there are targets with irregular status in the target information to be detected, the status standardization is irregular status and an irregular status alarm signal is generated; wherein, targets with irregular status include persons without helmets, persons without protective clothing and persons smoking.
6. The oilfield produced water transportation monitoring video detection method according to claim 1, characterized in that: The obtaining of behavior normative information according to each behavior information to be detected includes: When there is an irregular behavior in each behavior information to be detected, the behavior standardization information is irregular behavior and a irregular behavior alarm signal is generated; wherein, the irregular behavior includes produced water leakage and personnel falling behavior.
7. The oilfield produced water transportation monitoring video detection method according to claim 1, characterized in that: Also includes: The standardized inspection results of oilfield produced water transportation are visually displayed in the form of a web interface and / or a desktop application interface.
8. An oilfield produced water transportation monitoring video detection system, characterized in that: include: A data acquisition module, used to obtain monitoring videos of oilfield produced water transportation; A standard detection module is used to input the monitoring video of oilfield produced water transportation into the pre-trained operation standard detection and recognition model to obtain the standard detection result of oilfield produced water transportation; Among them, the operation specification detection and recognition model includes a target detection module, a behavior analysis module and a text information classification module which are connected in sequence; the target detection module is constructed based on the improved YOLOv8 target detection model, which is obtained by adding an attention mechanism to the backbone network of the YOLOv8 target detection model; the behavior analysis module is constructed based on the improved SlowFast behavior recognition model, which is obtained by replacing the loss function of the SlowFast behavior recognition model with the focal loss function Focal Loss; the target detection module is used to obtain the information of each target to be detected and the analysis scope of each behavior to be detected in the oilfield produced water transportation monitoring video, and to obtain state normative information according to the information of each target to be detected; the behavior analysis module is used to obtain the information of each behavior to be detected according to the information of each target to be detected and the analysis scope of each behavior to be detected, and to obtain behavior normative information according to the information of each behavior to be detected; the text information classification module is used to connect the information of each target to be detected and the information of each behavior to be detected into text and classify it to obtain process normative information.
9. A computer 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 computer program, the steps of the oilfield produced water transportation monitoring video detection method as described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the oilfield produced water transportation monitoring video detection method as described in any one of claims 1 to 7 are implemented.
Citation Information
Patent Citations
Oil field well site video target detection and identification method and system based on neural network
CN113191273A
Gas cylinder filling link illegal behavior identification method and system based on Internet of Things
CN115995119A
Industrial defect detection method based on deep network
CN117237325A
Target perception and personnel behavior posture detection method and system based on swimming pool environment
CN118053205A
Rehabilitation action recognition method and device and computer storage medium
CN118097778A
Cited By
Visual analysis-based crane crane operation risk detection method and system
CN121121646A
