Intelligent application management system and device

Abstract

The invention belongs to the field of oil depot management, and discloses an intelligent application management system and device, and the intelligent application management system comprises a personnel positioning management module, an intelligent inspection module, a video intelligent analysis module and a mobile application module. According to the method, different alarm levels are divided, response is carried out according to the emergency degree and potential risk of the situation, it is ensured that resources and attention are focused on the most needed place, abnormal behavior recognition based on deep learning and emotional state monitoring based on sound features can accurately evaluate the behavior and psychological state of personnel, and the accuracy of the abnormal behavior recognition is improved. Early identification of factors which may cause errors; by continuously updating the position of the target object in the continuous frames, movement of the object can be monitored in real time, it is ensured that the position of the object can be accurately known at any moment, potential unsafe behaviors or abnormal activities can be recognized in real time through tracking data, and early warning is given out in time.

Description

Technical Field

[0001] The present invention relates to the field of oil depot management, and in particular to an intelligent application management system and a device. Background Art

[0002] Jet fuel depots are facilities used to store and distribute aviation fuel. These depots are usually located at or near airports, and their main function is to ensure that airlines have access to the necessary fuel supplies to maintain normal flight operations.

[0003] The safety risks of aviation fuel depots mainly include fire, leakage, environmental pollution and the resulting personal injury or death. The highly flammable substances and complex oil transportation system inside the depot make strict safety measures and monitoring systems necessary. Management negligence, technical failures or operational errors may lead to serious safety accidents.

[0004] The intelligent application management of aviation fuel depot safety risks requires monitoring the status of personnel in the central control room, but the existing technology does not integrate emotional status and behavioral data for analysis, which may lead to incomplete early warning of potential problems and missed opportunities for early intervention. At the same time, without real-time tracking and analysis of the location and behavior of the target object, it may be impossible to discover and respond to potential safety factors in a timely manner.

[0005] Therefore, how to provide an intelligent application management system and device is a problem that needs to be solved urgently. Summary of the invention

[0006] The embodiment of the present invention provides an intelligent application management system and device to solve the problem that the prior art does not integrate and analyze emotional state and behavior data, which may lead to incomplete early warning of potential problems and miss the opportunity for early intervention. At the same time, it does not track and analyze the location and behavior of the target object in real time, which may make it impossible to discover and respond to potential safety factors in a timely manner.

[0007] In order to have a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended to be a general review, nor is it intended to identify key / important components or to delineate the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0008] According to a first aspect of an embodiment of the present invention, there is provided an intelligent application management system, which includes a personnel positioning management module, an intelligent inspection module, a video intelligent analysis module and a mobile application module.

[0009] Among them, the positioning management module is used to ensure the safety of personnel and vehicles inside the oil depot through positioning and real-time monitoring.

[0010] The intelligent inspection module is used to optimize the inspection process of the oil depot through information integration and real-time data interaction.

[0011] The video intelligent analysis module is used to manage the oil depot based on artificial intelligence technology and use video analysis technology to identify the trajectory of the target object.

[0012] The mobile application module is used for oil depot personnel to access the intelligent application management system based on mobile application services.

[0013] In one embodiment, the positioning management module includes a real-name entry and exit management module, an alarm management module, an alarm area management module, a real-time alarm display module and an emergency management module.

[0014] Among them, the real-name entry and exit management module is used to manage the entry and exit of people and vehicles through the real-name system;

[0015] Alarm management module, used to monitor and manage various security alarms;

[0016] The alarm area management module is used to draw the alarm area through electronic fence technology and monitor the high-risk operation area;

[0017] The alarm real-time display module is used to display the alarm information of each monitoring area in real time and realize the linkage with the real-time video;

[0018] The emergency management module is used for on-site personnel to send out distress signals and locate people in danger.

[0019] In one embodiment, the intelligent inspection module includes a line management module and a track recording module.

[0020] The line management module is used to set inspection lines, allocate inspection points, manage inspection points, and manage the equipment information bound to the inspection points;

[0021] The trajectory recording module is used to record the inspector's action trajectory, support trajectory playback, and record the data collected during the inspection process.

[0022] In one embodiment, the video intelligent analysis module includes a control room monitoring module and a target tracking module.

[0023] The central control room monitoring module is used to identify abnormal behaviors and emotional states of personnel in the central control room and implement graded alarms;

[0024] The target tracking module is used to track the trajectory of the target object based on video analysis technology.

[0025] In one embodiment, identifying abnormal behavior and emotional state of personnel in the central control room and implementing graded alarms includes:

[0026] Classify alarm levels, including low-level alarm, medium-level alarm and high-level alarm;

[0027] Acquire abnormal behavior recognition data based on deep learning models;

[0028] Identify the emotional state of control room personnel based on voice characteristics;

[0029] The emotional state recognition data and abnormal behavior recognition data are integrated, and based on the logic algorithm, the alarm level is determined according to the severity of the emotional state and abnormal behavior.

[0030] In one embodiment, identifying the emotional state of the personnel in the control room based on the sound characteristics includes:

[0031] Collect voice samples of the control room personnel in normal and different emotional states, and pre-process the voice signals:

[0032] Calculate the energy of each frame, classify the frames according to the energy value, and calculate the time proportion of each type of frame in the total number of frames;

[0033] Extract short-time average frequency, short-time mean square error frequency and formant frequency as rhythmic features, and construct a three-dimensional sound feature dataset:

[0034] Combine time scale and fusion prosodic features into a three-dimensional sound feature dataset;

[0035] A clustering algorithm is used to find cluster centers in a three-dimensional sound feature dataset, and a Gaussian mixture model is trained using the cluster centers.

[0036] The Gaussian mixture model is initialized using the sound data of the control room personnel in a normal emotional state, and the optimal cluster center and the corresponding Gaussian mixture model are determined as the benchmark;

[0037] The voices of the personnel in the control room are collected in real time and input into the baseline model and other Gaussian mixture models to obtain the probabilities of different emotional states.

[0038] In one embodiment, the emotional state recognition data and the abnormal behavior recognition data are integrated, and based on a logic algorithm, the alarm level is determined according to the severity of the emotional state and the abnormal behavior, including:

[0039] Obtain key features in emotional state recognition data and abnormal behavior recognition data;

[0040] Define the quantitative indicators of key features and calculate the comprehensive score based on the quantitative indicators and corresponding weights;

[0041] Set different alarm level thresholds, and determine the alarm level based on the comparison between the comprehensive score and the preset alarm level threshold.

[0042] In one embodiment, tracking the trajectory of a target object based on video analysis technology includes:

[0043] Collect video data, preprocess the video stream, and use computer vision algorithms to identify target objects in the video;

[0044] Extract the features of the target object and use the tracking algorithm to update the position and state of the target object in consecutive frames;

[0045] The position information of the target object in consecutive frames is connected in series to form the trajectory of the target object.

[0046] In one embodiment, using a computer vision algorithm to identify a target object in a video includes:

[0047] Collect images of existing target objects to form a data set, use the convolutional neural network of the YOLO model to extract features from the images, and form feature maps of different scales;

[0048] Use the labeled data set to train the YOLO model, and use the convolutional neural network to learn the features of the target object image to achieve multi-scale detection;

[0049] Input the video sequence into the trained YOLO model to complete real-time target object recognition.

[0050] According to a second aspect of an embodiment of the present invention, there is provided an intelligent application management device. The device comprises:

[0051] processor;

[0052] a memory for storing processor-executable instructions;

[0053] Wherein, the processor is configured to:

[0054] Ensure the safety of personnel and vehicles inside the oil depot through positioning and real-time monitoring;

[0055] Optimize the inspection process of oil depots through information integration and real-time data interaction;

[0056] Manage oil depots based on artificial intelligence technology and use video analysis technology to identify the trajectory of target objects;

[0057] Based on mobile application services, oil depot personnel access the intelligent application management system.

[0058] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0059] (1) The present invention monitors the positions of personnel and vehicles in the oil depot in real time, can quickly respond to potential safety issues or violations, and ensures that inspection tasks are carried out according to predetermined standards and time through automated data integration and real-time information interaction, thereby reducing human errors.

[0060] (2) The present invention divides alarm levels into different levels and responds according to the urgency and potential risks of the situation, ensuring that resources and attention are focused where they are most needed. Abnormal behavior recognition based on deep learning and emotional state monitoring based on sound features can accurately assess the behavior and psychological state of personnel and identify factors that may lead to errors at an early stage. Fusion of emotional state recognition data and abnormal behavior recognition data makes monitoring more comprehensive.

[0061] (3) The present invention can monitor the movement of the object in real time by continuously updating the position of the target object in consecutive frames, ensuring that the position of the object can be accurately known at any time. The tracking data can identify potential unsafe behaviors or abnormal activities in real time and issue early warnings in a timely manner. At the same time, the trajectory data can be used to analyze the flow and activity patterns during peak hours, helping management to make more reasonable personnel and security resource allocation decisions.

[0062] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0064] Figure 1 It is a principle block diagram of an intelligent application management system according to an exemplary embodiment.

[0065] Reference numerals:

[0066] 101. Personnel positioning management module; 103. Intelligent inspection module; 105. Video intelligent analysis module; 107. Mobile application module. DETAILED DESCRIPTION

[0067] The following description and accompanying drawings fully illustrate the specific embodiments of this article so that those skilled in the art can practice them. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. The scope of the embodiments of this article includes the entire scope of the claims, as well as all available equivalents of the claims. Herein, the terms "first", "second", etc. are only used to distinguish one element from another, without requiring or implying any actual relationship or order between these elements. In fact, the first element can also be called the second element, and vice versa. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the structure, device or equipment including a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also include elements inherent to such structure, device or equipment. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the structure, device or equipment including the elements. Each embodiment is described in a progressive manner herein, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the embodiments can be referred to each other.

[0068] The terms "longitudinal", "lateral", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. in this document indicate the orientation or position relationship based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing this document and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In the description of this document, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a mechanical connection or an electrical connection, it can also be the internal communication of two elements, it can be a direct connection, or it can be an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0069] As used herein, the term "plurality" means two or more than two, unless otherwise specified.

[0070] In this document, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0071] In this article, the term "and / or" is a description of the association relationship between objects, indicating that three relationships may exist. For example, A and / or B means: A or B, or, A and B.

[0072] It should be understood that, although the various steps in the flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the figure may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0073] Each module in the device or system of the present application can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above modules.

[0074] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0075] Figure 1 An embodiment of an intelligent application management system of the present invention is shown.

[0076] In this optional embodiment, the intelligent application management system includes:

[0077] Personnel positioning management module 101, intelligent inspection module 103, video intelligent analysis module 105 and mobile application module 107.

[0078] Among them, the positioning management module 101 is used to ensure the safety of personnel and vehicles inside the oil depot through positioning and real-time monitoring, optimize human resource allocation, and improve safety efficiency.

[0079] In one embodiment, preferably, the positioning management module 101 includes a real-name entry and exit management module, an alarm management module, an alarm area management module, an alarm real-time display module and an emergency management module;

[0080] The real-name entry and exit management module is used to manage the entry and exit of personnel and vehicles by real-name system, using identity authentication technology (such as card swiping, fingerprint, face recognition) to ensure that only authorized personnel and vehicles can enter key areas. At the same time, the number and area of ​​​​entry and exit personnel are displayed in real time, and detailed query and statistics functions are provided.

[0081] The alarm management module is used to monitor and manage various safety alarms, including personnel leaving their posts, visiting other posts, and overcrowding alarms, as well as vehicle speeding and deviation from the route, etc. It supports multiple alarm methods such as sound and light, SMS prompts, and one-click call for help, realizing closed-loop processing of alarm events.

[0082] The alarm area management module is used to draw specific alarm areas through electronic fence technology to conduct special monitoring of high-risk operation areas. It supports online headcount and timeout alarm functions to enhance the safety management of the operation area.

[0083] The real-time alarm display module is used to combine GIS technology to display the alarm information of each monitoring area in real time, such as one-click distress alarms and regional personnel alarms, to achieve linkage with real-time video and improve monitoring efficiency and accuracy.

[0084] The emergency management module is used to enable on-site personnel to quickly send out a distress signal through a one-touch distress button in an emergency, and the management personnel can use video surveillance and personnel positioning functions to quickly locate the people in danger and carry out effective on-site emergency response.

[0085] The intelligent inspection module 103 is used to optimize the inspection process of the oil depot through information integration and real-time data interaction.

[0086] In one embodiment, preferably, the intelligent inspection module 103 includes a line management module and a track recording module.

[0087] The line management module is used to set inspection lines, assign inspection points, manage inspection points, and manage the equipment information bound to the inspection points. Administrators can assign specific lines to inspectors to systematically manage the inspection process. Inspectors can obtain relevant equipment information by scanning the QR code on the device to facilitate the execution of inspection tasks.

[0088] The trajectory recording module is used to record the inspector's action trajectory and supports trajectory playback, which helps to monitor whether the inspector's actual action path is consistent with the plan. At the same time, it records the data collected during the inspection, such as pictures and videos, which are used for abnormal situations.

[0089] The video intelligent analysis module 105 is used to manage the oil depot based on artificial intelligence technology and to identify the trajectory of the target object using video analysis technology.

[0090] In one embodiment, preferably, the video intelligent analysis module 105 includes a central control room monitoring module and a target tracking module.

[0091] The central control room monitoring module is used to identify abnormal (off-duty, sleeping on duty, etc.) behavior and emotional state of central control room personnel and implement graded alarms.

[0092] The target tracking module is used to track the trajectory of the target object based on video analysis technology.

[0093] In one embodiment, preferably, identifying abnormal behavior and emotional state of personnel in the central control room and implementing graded alarms includes:

[0094] The alarm levels are divided into low-level alarms, medium-level alarms and high-level alarms.

[0095] Acquire abnormal behavior recognition data based on deep learning models.

[0096] Identify the emotional state of control room personnel based on sound characteristics.

[0097] The emotional state recognition data and abnormal behavior recognition data are integrated, and based on the logic algorithm, the alarm level is determined according to the severity of the emotional state and abnormal behavior.

[0098] Among them, the abnormal behavior recognition data obtained based on the deep learning model includes:

[0099] Define what constitutes abnormal behavior, for example, long periods of stillness, sudden rapid movements, activity outside of work hours, etc. Collect video data, ensuring that samples of normal and abnormal behavior are covered, and collect timestamp data associated with the behavior.

[0100] The collected video data is annotated to mark normal and abnormal behaviors, as well as the time when the behaviors occur and how long they last.

[0101] Use deep learning models to automatically extract features from video frames, or manually design features and use model learning. Choose a deep learning model suitable for behavior recognition, such as convolutional neural networks, recurrent neural networks, long short-term memory networks, etc.

[0102] Train a deep learning model using labeled data and tune model parameters for optimal performance. Train an additional model or use an existing model to predict the duration of abnormal behavior. Use test data to evaluate model performance metrics such as accuracy, recall, and F1 score.

[0103] Deploy the trained deep learning model to the real-time monitoring system to analyze the real-time video stream, identify abnormal behaviors in real time, and predict their duration.

[0104] In one embodiment, preferably, identifying the emotional state of the personnel in the central control room based on the sound characteristics includes:

[0105] Collect voice samples of control room personnel in normal and different emotional states, and pre-process the sound signals.

[0106] Calculate the energy of each frame, classify the frames according to the energy value, and calculate the time proportion of each type of frame in the total number of frames.

[0107] The short-time average frequency, short-time mean square error frequency and formant frequency are extracted as rhythmic features, and a three-dimensional sound feature dataset is constructed.

[0108] Combining time scale and fused prosodic features into a three-dimensional sound feature dataset.

[0109] A clustering algorithm is used to find cluster centers in the three-dimensional sound feature dataset. Each cluster center represents a specific emotional state. The cluster centers are used to train a Gaussian mixture model. Each cluster center corresponds to a Gaussian mixture model, which is used to identify the corresponding emotional state.

[0110] The Gaussian mixture model is initialized using the sound data of the control room personnel in a normal emotional state, and the optimal clustering center and the corresponding Gaussian mixture model are determined as a benchmark.

[0111] The voices of the personnel in the control room are collected in real time and input into the baseline model and other Gaussian mixture models to obtain the probabilities of different emotional states.

[0112] In one embodiment, preferably, the emotional state recognition data and the abnormal behavior recognition data are integrated, and based on a logic algorithm, the alarm level is determined according to the severity of the emotional state and the abnormal behavior, including:

[0113] Obtain key features in emotional state recognition data and abnormal behavior recognition data.

[0114] Among them, key features such as emotion probability, emotion type, etc. are extracted from the emotion state recognition data.

[0115] Extract key features from abnormal behavior recognition data, such as behavior type, behavior frequency, behavior duration, etc.

[0116] Define quantitative indicators of key features and calculate comprehensive scores based on the quantitative indicators and corresponding weights.

[0117] Set different alarm level thresholds, and determine the alarm level based on the comparison between the comprehensive score and the preset alarm level threshold.

[0118] In one embodiment, preferably, implementing trajectory tracking of a target object based on video analysis technology includes:

[0119] Collect video data, perform preprocessing such as denoising and contrast adjustment on the video stream to improve image quality, and use computer vision algorithms to identify target objects in the video;

[0120] Extract the features of the target object, such as color, shape, texture, etc., for subsequent tracking analysis, and use tracking algorithms such as Kalman filtering, correlation tracking, and deep learning-based tracking to update the position and state of the target object in consecutive frames;

[0121] The position information of the target object in consecutive frames is connected in series to form the trajectory of the target object.

[0122] In one embodiment, preferably, using a computer vision algorithm to identify a target object in a video includes:

[0123] Collect images of existing target objects to form a data set, use the convolutional neural network of the YOLO model to extract features from the images, and form feature maps of different scales.

[0124] The YOLO model is trained using annotated datasets so that it can recognize images of target objects and learn the features of target object images through convolutional neural networks to achieve multi-scale detection.

[0125] Input the video sequence into the trained YOLO model to complete real-time target object recognition.

[0126] The mobile application module 4 is used for oil depot personnel to access the intelligent application management system based on the mobile application service (Application), thereby enhancing the convenience of communication and operation.

[0127] The mobile application module 107 allows oil depot staff to access the intelligent application management system through smartphones or tablets at any location. In-application communication tools (such as message push, instant chat, etc.) can enhance communication between oil depot personnel. Mobile applications configure real-time alerts and notifications.

[0128] An embodiment of an intelligent application management device of the present invention provides an intelligent application management device, including:

[0129] processor;

[0130] a memory for storing processor-executable instructions;

[0131] Wherein, the processor is configured to:

[0132] Ensure the safety of personnel and vehicles inside the oil depot through positioning and real-time monitoring;

[0133] Optimize the inspection process of oil depots through information integration and real-time data interaction;

[0134] Manage oil depots based on artificial intelligence technology and use video analysis technology to identify the trajectory of target objects;

[0135] Based on mobile application services, oil depot personnel access the intelligent application management system.

[0136] The present invention is not limited to the structures which have been described above and shown in the drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. An intelligent application management system, characterized in that: The intelligent application management system includes a personnel positioning management module, an intelligent inspection module, a video intelligent analysis module and a mobile application module; The positioning management module is used to ensure the safety of personnel and vehicles inside the oil depot through positioning and real-time monitoring; The intelligent inspection module is used to optimize the inspection process of the oil depot through information integration and real-time data interaction; The video intelligent analysis module is used to manage the oil depot based on artificial intelligence technology and identify the trajectory of the target object using video analysis technology; The mobile application module is used for oil depot personnel to access the intelligent application management system based on mobile application services.

2. An intelligent application management system according to claim 1, characterized in that: The positioning management module includes a real-name entry and exit management module, an alarm management module, an alarm area management module, a real-time alarm display module and an emergency management module; The real-name entry and exit management module is used to manage the entry and exit of people and vehicles by real-name system; The alarm management module is used to monitor and manage various security alarms; The alarm area management module is used to draw the alarm area through the electronic fence technology and monitor the high-risk operation area; The alarm real-time display module is used to display the alarm information of each monitoring area in real time and realize linkage with real-time video; The emergency management module is used for on-site personnel to send out distress signals and locate people in danger.

3. The intelligent application management system according to claim 1, characterized in that: The intelligent inspection module includes a line management module and a track recording module; The line management module is used to set inspection lines, allocate inspection points, manage inspection points, and manage equipment information bound to inspection points; The track recording module is used to record the action track of the inspector, support track playback, and record the data collected during the inspection process.

4. The intelligent application management system according to claim 1, characterized in that: The video intelligent analysis module includes a central control room monitoring module and a target tracking module; The control room monitoring module is used to identify abnormal behaviors and emotional states of personnel in the control room and implement graded alarms; The target tracking module is used to track the trajectory of the target object based on video analysis technology.

5. An intelligent application management system according to claim 4, characterized in that: The identification of abnormal behavior and emotional state of personnel in the central control room and the implementation of graded alarms include: Classify alarm levels, including low-level alarm, medium-level alarm and high-level alarm; Acquire abnormal behavior recognition data based on deep learning models; Identify the emotional state of control room personnel based on voice characteristics; The emotional state recognition data and abnormal behavior recognition data are integrated, and based on the logic algorithm, the alarm level is determined according to the severity of the emotional state and abnormal behavior.

6. An intelligent application management system according to claim 5, characterized in that: The method of identifying the emotional state of the personnel in the control room based on the sound characteristics includes: Collect voice samples of the control room personnel in normal and different emotional states, and pre-process the voice signals: Calculate the energy of each frame, classify the frames according to the energy value, and calculate the time proportion of each type of frame in the total number of frames; Extract short-time average frequency, short-time mean square error frequency and formant frequency as rhythmic features, and construct a three-dimensional sound feature dataset: Combine time scale and fusion prosodic features into a three-dimensional sound feature dataset; A clustering algorithm is used to find cluster centers in a three-dimensional sound feature dataset, and a Gaussian mixture model is trained using the cluster centers. The Gaussian mixture model is initialized using the sound data of the control room personnel in a normal emotional state, and the optimal cluster center and the corresponding Gaussian mixture model are determined as the benchmark; The voices of the personnel in the control room are collected in real time and input into the baseline model and other Gaussian mixture models to obtain the probabilities of different emotional states.

7. An intelligent application management system according to claim 6, characterized in that: The step of fusing the emotional state recognition data and the abnormal behavior recognition data and determining the alarm level based on the severity of the emotional state and the abnormal behavior based on a logic algorithm includes: Obtain key features in emotional state recognition data and abnormal behavior recognition data; Define the quantitative indicators of key features and calculate the comprehensive score based on the quantitative indicators and corresponding weights; Set different alarm level thresholds, and determine the alarm level based on the comparison between the comprehensive score and the preset alarm level threshold.

8. An intelligent application management system according to claim 7, characterized in that: The tracking of the target object based on the video analysis technology includes: Collect video data, preprocess the video stream, and use computer vision algorithms to identify target objects in the video; Extract the features of the target object and use the tracking algorithm to update the position and state of the target object in consecutive frames; The position information of the target object in consecutive frames is connected in series to form the trajectory of the target object.

9. An intelligent application management system according to claim 8, characterized in that: The use of a computer vision algorithm to identify a target object in a video includes: Collect images of existing target objects to form a data set, use the convolutional neural network of the YOLO model to extract features from the images, and form feature maps of different scales; Use the labeled data set to train the YOLO model, and use the convolutional neural network to learn the features of the target object image to achieve multi-scale detection; Input the video sequence into the trained YOLO model to complete real-time target object recognition.

10. An intelligent application management device, characterized in that: The device comprises: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: Ensure the safety of personnel and vehicles inside the oil depot through positioning and real-time monitoring; Optimize the inspection process of oil depots through information integration and real-time data interaction; Manage oil depots based on artificial intelligence technology and use video analysis technology to identify the trajectory of target objects; Based on mobile application services, oil depot personnel access the intelligent application management system.

Images