Chemical production safety intelligent management system based on video monitoring

Abstract

The invention relates to the technical field of chemical production management, in particular to a chemical production safety intelligent management system based on video monitoring, which comprises a video stream optimization module, a behavior analysis and evaluation module, a real-time risk management module, a monitoring process optimization module, a behavior mode updating module and an emergency response correction module. According to the invention, by optimizing the quality of the video stream, the high definition and the real-time performance of the monitoring content are ensured, so that the system can more effectively identify abnormal conditions in production, accurately identify the deviation between the operation and the safety standard through meticulous behavior tracking and quantitative analysis, and adjust the deviation in time, thereby improving the production efficiency. Therefore, the normalization of the operation behavior is accurately evaluated, the behavior deviating from the standard is corrected in time, the effectiveness of safety management is improved, and the early warning capability of potential risks is improved through automatic safety level classification and early warning threshold adjustment in real-time risk management, so that the safety response is more flexible, and the probability of accidents is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical production management, and particularly to an intelligent management system for chemical production safety based on video surveillance. Background Art

[0002] Chemical production management is an important technical field in the chemical industry, involving multiple aspects of chemical production, processing, and control. The main objectives are to improve production efficiency, ensure product quality, optimize resource utilization, and minimize environmental impact. With the progress of technology, especially the development of information technology and automation technology, chemical production management is gradually moving towards a more intelligent and automated direction.

[0003] Among them, the intelligent management system for chemical production safety based on video surveillance focuses on using video surveillance technology to enhance the safety management of chemical production. The system uses cameras installed in key production areas to monitor physical and chemical reactions during the production process in real time, as well as the safety behavior of workers. Through the analysis of video streams, the system can automatically detect potential safety hazards, such as abnormal operation of equipment, leakage, or entry of unauthorized personnel into restricted areas. It can not only improve the response speed to emergencies, reduce the risk of accidents, but also provide decision-making support for future safety management through historical data analysis.

[0004] Traditional systems have limitations in image processing. For example, it is difficult to adjust in real time for changes in lighting and dynamic capture in a rapidly changing production environment, which affects the accuracy of anomaly recognition. There is a lack of automated safety level classification and threshold adjustment in the risk assessment process, resulting in the warning system being unable to optimize in real time according to the actual risk, causing delays or false alarms when dealing with emergencies, increasing the complexity and resource consumption of accident handling. These limitations are particularly prominent in dealing with high-risk chemical production processes and often lead to safety accidents due to untimely or inaccurate responses. Summary of the Invention

[0005] In order to solve the technical problems existing in the prior art, such as limitations in image processing, for example, it is difficult to adjust in real time for changes in lighting and dynamic capture in a rapidly changing production environment, which affects the accuracy of anomaly recognition. There is a lack of automated safety level classification and threshold adjustment in the risk assessment process, resulting in the warning system being unable to optimize in real time according to the actual risk, causing delays or false alarms when dealing with emergencies, increasing the complexity and resource consumption of accident handling. These limitations are particularly prominent in dealing with high-risk chemical production processes and often lead to safety accidents due to untimely or inaccurate responses, the embodiments of the present invention provide an intelligent management system for chemical production safety based on video surveillance. The technical solution is as follows:

[0006] On the one hand, a chemical production safety intelligent management system based on video surveillance is provided, including:

[0007] The video stream optimization module is based on the real-time video stream of chemical production captured by the camera, performs image resolution and frame rate adjustment, eliminates the influence of background noise and light changes, decodes and compresses the signal, and then corrects color distortion and motion blur to obtain an optimized video stream quality.

[0008] The behavior analysis and evaluation module is based on the optimized video stream quality, tracks the behavior changes between key frames, identifies the deviations from the standard operations, and evaluates the normativity of the operation behavior based on the deviations to obtain the behavior deviation index.

[0009] The real-time risk management module is based on the behavior deviation index, identifies the behavior patterns deviating from the standard operations, automatically classifies the safety levels, calculates and predicts the occurrence probability of multi-level risks, and automatically adjusts the warning threshold to obtain the risk level response information.

[0010] The monitoring process optimization module is based on the risk level response information, distributes the warning notifications to the monitoring center and the on-site operation area, monitors the personnel status in the on-site operation area, adjusts the monitoring parameters according to the personnel feedback information, and evaluates the warning reaction effect to obtain the warning effect evaluation result.

[0011] The behavior pattern update module is based on the warning effect evaluation result, continuously monitors the behavior pattern characteristics in the video stream, compares with the historical behavior data, and identifies new behavior trends or behavior changes to obtain an updated behavior database.

[0012] The emergency response correction module is based on the updated behavior database, provides behavior correction guidance to the operators, formulates emergency shutdown and intervention measures for the monitored risk behaviors, and obtains the accident prevention and control management log.

[0013] On the other hand, the optimized video stream quality includes signal integration, dynamic range, and frame integrity. The behavior deviation index includes deviation intensity, consistency index, and behavior deviation frequency. The risk level response information includes risk discrimination threshold, response level classification, and warning signal intensity. The warning effect evaluation result includes response effectiveness, reaction adjustment time, and monitoring efficiency. The updated behavior database is specifically the behavior pattern recognition result, pattern change record, and behavior sequence analysis result.

[0014] On the other hand, the video stream optimization module includes an image quality adjustment sub-module and an image stabilization sub-module.

[0015] The image quality adjustment sub-module adjusts the resolution and frame rate of the image based on the real-time video stream of chemical production captured by the camera, matches the environmental requirements of differential monitoring, and optimizes the influence of background noise and light changes to obtain a video quality optimization record;

[0016] The signal processing sub-module decodes and compresses the signals of the video quality optimization record, adjusts the transmission parameters and storage requirements of the data, corrects color distortion and motion blur, and optimizes the clarity of the monitoring screen to obtain an optimized video stream quality.

[0017] On the other hand, the behavior analysis and evaluation module includes a behavior tracking sub-module, a deviation analysis sub-module, and a norm evaluation sub-module;

[0018] The behavior tracking sub-module monitors the behaviors in the video frame sequence based on the optimized video stream quality, captures the behavior changes between key frames, marks each key frame with a timestamp, and associates the behavior events with the operation areas to refine the corresponding relationship between time and space data, and obtains behavior tracking mapping data;

[0019] The deviation analysis sub-module makes a frame-by-frame comparison between the operator's behavior and the standard behavior based on the behavior tracking mapping data, identifies behavior deviations, determines the degree of deviation, and counts the occurrence frequency and time distribution of the deviation behaviors to obtain a behavior deviation metric;

[0020] The norm evaluation sub-module performs a behavior compliance check based on the behavior deviation metric, compares it with the safety operation standard, evaluates the compliance range of the behavior deviation, and quantifies the potential impact of the deviation on the safety baseline to obtain a behavior deviation index.

[0021] On the other hand, the frame-by-frame comparison between the operator's behavior and the standard behavior, identifying behavior deviations, and determining the degree of deviation, uses the formula:

[0022]

[0023] Calculate the behavior deviation metric value D, count the occurrence frequency and time distribution of the deviation behaviors to obtain a behavior deviation metric, where α k represents the key weight of the k-th frame, O k represents the operator's behavior data in the k-th frame, S k represents the standard behavior data in the k-th frame, and m is the total number of frames.

[0024] On the other hand, the real-time risk management module includes a behavior pattern recognition sub-module, a safety level classification sub-module, and a warning adjustment sub-module;

[0025] Based on the behavior deviation index, the behavior pattern recognition sub-module identifies the behavior patterns that deviate from the standard operations, classifies them according to the degree of deviation of the behavior from the standard, determines the behaviors that belong to the deviation from the standard operations, and obtains a behavior pattern recognition list;

[0026] Based on the behavior pattern recognition list, the security level classification sub-module classifies the monitored behavior patterns according to the risk severity, evaluates the potential threat of each behavior pattern to security, and divides the security level according to the risk severity to obtain risk level distribution information;

[0027] Based on the risk level distribution information, the early warning adjustment sub-module automatically adjusts the early warning threshold, and dynamically adjusts the early warning response level according to the changing security risk data to obtain risk level response information.

[0028] On the other hand, classifying the monitored behavior patterns according to the risk severity, evaluating the potential threat of each behavior pattern to security, using the formula:

[0029]

[0030] And dividing the security level according to the risk severity to obtain risk level distribution information, where R i represents the risk level of the i-th behavior pattern, β j represents the risk coefficient of the j-th behavior feature, f(B j ) represents the risk function value of the j-th behavior feature, and n represents the total number of behavior features.

[0031] On the other hand, the monitoring process optimization module includes an early warning information distribution sub-module, a monitoring optimization sub-module, and an early warning effect evaluation sub-module;

[0032] Based on the risk level response information, the early warning information distribution sub-module extracts the early warning content, preferentially transmits the early warning notifications of high risk levels to the monitoring center, determines the sending status of the notification information, records the sending time and the received feedback information, and obtains the notification transmission record;

[0033] Based on the notification transmission record, the monitoring optimization sub-module analyzes the response behavior data of the on-site personnel, adjusts the direction and focal length range of the monitoring camera, monitors the dynamic changes of the on-site operation area in real time, corrects the position of the monitoring blind area, and generates monitoring adjustment details;

[0034] Based on the monitoring adjustment details, the early warning effect evaluation sub-module screens the received early warning feedback data, counts the response time and coverage range of different early warning levels, compares them with the processing situation of the early warning information and the standard processing requirements, evaluates the early warning reaction effect, and obtains the early warning effect evaluation result.

[0035] On the other hand, the behavior pattern update module includes a behavior analysis sub-module, a behavior comparison sub-module, and a database update sub-module;

[0036] Based on the early warning effect evaluation result, the behavior analysis sub-module performs real-time analysis on the behavior data captured in the video stream, extracts key behavior features, separates and marks key behavior changes, and obtains a behavior feature extraction result;

[0037] Based on the behavior feature extraction result, the behavior comparison sub-module compares the current behavior data with the behavior patterns stored in the database, identifies new or changed behavior trends, and obtains behavior trend comparison information;

[0038] Based on the behavior trend comparison information, the database update sub-module incorporates the newly mined behavior trends and key data points into the database and optimizes the data structure to obtain an updated behavior database.

[0039] On the other hand, the emergency response correction module includes a behavior guidance sub-module and an accident management sub-module;

[0040] Based on the updated behavior database, the behavior guidance sub-module provides behavior guidance to the operators, analyzes unsafe behavior patterns, and through simulation training and behavior correction, refines the operators' understanding results of risk behaviors to obtain behavior correction guidance data;

[0041] Based on the behavior correction guidance data, the accident management sub-module formulates emergency shutdown and intervention measures for high-risk behaviors, adjusts the operation process and matches potential safety threats, optimizes the accident occurrence probability and intervention effectiveness, and obtains an accident prevention and control management log.

[0042] The beneficial effects brought by the technical solutions provided by the embodiments of the present invention at least include:

[0043] By optimizing the quality of the video stream, the high definition and real-time nature of the monitored content are ensured, enabling the system to more effectively identify abnormal situations in production. Through detailed behavior tracking and quantitative analysis, the deviation between operations and safety standards can be accurately identified and adjusted in a timely manner, thereby precisely evaluating the standardization of operation behaviors, promptly correcting behaviors that deviate from the standards, improving the effectiveness of safety management. Real-time risk management enhances the early warning ability for potential hazards through automated safety level classification and warning threshold adjustment, making the safety response more flexible, significantly reducing the probability of accidents. The real-time evaluation of the comprehensive monitoring and early warning effects further enhances the dynamic adaptability of risk management, ensuring the simultaneous improvement of production safety and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0045] Figure 1 It is a schematic diagram of the system of the present invention;

[0046] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0047] Figure 3 It is a flowchart of the video stream optimization module of the present invention;

[0048] Figure 4 It is a flowchart of the behavior analysis and evaluation module of the present invention;

[0049] Figure 5 It is a flowchart of the real-time risk management module of the present invention;

[0050] Figure 6 It is a flowchart of the monitoring process optimization module of the present invention;

[0051] Figure 7 It is a flowchart of the behavior pattern update module of the present invention;

[0052] Figure 8 It is a flowchart of the emergency response and correction module of the present invention. Detailed implementation manners

[0053] The following will describe the technical solutions in the present invention in conjunction with the drawings.

[0054] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly, the use of the word "example" is intended to present concepts in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either of the two can be selected.

[0055] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when not emphasizing their differences, the meanings they express are the same. "(of)", "corresponding" and "corresponding" can sometimes be used interchangeably. It should be noted that when not emphasizing their differences, the meanings they express are the same.

[0056] In the embodiments of the present invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0057] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0058] The embodiments of the present invention provide an intelligent management system for chemical production safety based on video surveillance, as Figure 1 shown. The system includes:

[0059] The video stream optimization module is based on the real-time video stream of chemical production captured by the camera, performs image resolution and frame rate adjustment, eliminates the influence of background noise and light changes, decodes and compresses the signal, and then corrects color distortion and motion blur to obtain optimized video stream quality;

[0060] The behavior analysis and evaluation module is based on the optimized video stream quality, tracks the behavior changes between key frames, identifies the deviation from the standard operation by quantitatively analyzing the difference between frames, and evaluates the normativeness of the operation behavior based on the deviation to obtain a behavior deviation index;

[0061] The real-time risk management module is based on the behavior deviation index, identifies the behavior patterns deviating from the standard operation, automatically classifies the safety levels according to the deviation degree, calculates and predicts the occurrence probability of multi-level risks, and automatically adjusts the warning threshold to obtain risk level response information;

[0062] The monitoring process optimization module is based on the risk level response information, distributes the warning notifications to the monitoring center and the on-site operation area, monitors the personnel status in the on-site operation area, adjusts the monitoring parameters according to the personnel feedback information, and evaluates the warning response effect to obtain a warning effect evaluation result;

[0063] The behavior pattern update module is based on the warning effect evaluation result, continuously monitors the behavior pattern characteristics in the video stream, compares with the historical behavior data, and identifies new behavior trends or behavior changes to obtain an updated behavior database;

[0064] The emergency response and correction module is based on the updated behavior database, provides behavior correction guidance to the operators, formulates emergency shutdown and intervention measures for the monitored risk behaviors, and optimizes the accident occurrence and potential risk probability to obtain an accident prevention and control management log.

[0065] The optimized video stream quality includes signal integration, dynamic range, and frame integrity. The behavior deviation indicators include deviation intensity, consistency index, and behavior deviation frequency. The risk level response information includes risk discrimination threshold, response level classification, and warning signal intensity. The warning effect evaluation results include response effectiveness, reaction adjustment time, and monitoring efficiency. The updated behavior database specifically includes behavior pattern recognition results, pattern change records, and behavior sequence analysis results.

[0066] As Figure 2 and Figure 3 shown, the video stream optimization module includes an image quality adjustment sub-module and an image stabilization sub-module;

[0067] Based on the real-time video stream of chemical production captured by the camera, the image quality adjustment sub-module adjusts the resolution and frame rate of the image, matches the environmental requirements of differential monitoring, and optimizes the effects of background noise and light changes to obtain video quality optimization records;

[0068] The video data is connected to the processing module through an interface. For the adjustment of image resolution, the clarity is adjusted by dynamically setting the specific parameters of the resolution. At the same time, for the video frame rate, the time interval of the frame sequence is adjusted to adapt to the needs of different environments. Then, the noise processing module is used to remove the noise in the video background, and unnecessary high-frequency noise and spot effects are filtered out through iterative calculation. At the same time, the light change is gradually corrected, the brightness change of the pixels is limited and the abnormal parts are repaired. Finally, the optimized video stream is saved as a record file for subsequent operations.

[0069] The signal processing sub-module decodes and compresses the signals of the video quality optimization records, adjusts the transmission parameters and storage requirements of the data, corrects color distortion and motion blur, and optimizes the clarity of the monitoring screen to obtain the optimized video stream quality;

[0070] The signal features in the video are extracted through standard decoding. The color signal is decomposed into basic channel information and abnormal data that does not meet the normal range is cleared. Then, compression technology is used to reduce the storage space requirements, and the compression parameters are adjusted to reduce the file size. At the same time, for the color distortion problem in the picture, the color display is gradually corrected by comparing the color deviation data of the video signal. Further, the motion blur problem in the picture is optimized, and the blurred part is compensated and repaired by adjusting the pixel position information of each frame of the picture to generate an optimized clear video stream.

[0071] As Figure 2 and Figure 4 shown, the behavior analysis and evaluation module includes a behavior tracking sub-module, a deviation analysis sub-module, and a normativity evaluation sub-module;

[0072] Based on the optimized video stream quality, the behavior tracking sub-module monitors the behaviors in the video frame sequence, captures the behavior changes between key frames, marks each key frame with a timestamp, associates the behavior events with the operation areas, refines the corresponding relationship between time and space data, and obtains the behavior tracking mapping data;

[0073] Parse the video frames sequentially, extract the key frames among them, set a frame difference threshold according to the feature areas with significant changes between frames to determine which frames are key frames, then mark the behavior characteristics of the key frames, use timestamps to record the occurrence time of each key frame, and at the same time identify the specific action patterns and spatial areas involved in each frame of the picture, divide the actions into multiple sub-categories, record the corresponding relationship between the sub-categories and the spatial positions. Subsequently, by constructing a time and space mapping table, record the occurrence time of each behavior event and the specific spatial operation area, and check the generated mapping data to ensure the completeness of the time record, the uniqueness and accuracy of the spatial area, and obtain the behavior tracking mapping data that can reflect the corresponding relationship between behavior and operation area.

[0074] Based on the behavior tracking mapping data, the deviation analysis sub-module compares the operator's behavior with the standard behavior frame by frame, identifies the behavior deviation, determines the degree of deviation, and counts the occurrence frequency and time distribution of the deviated behavior to obtain the behavior deviation metric;

[0075] Compare the actually collected operator's behavior with the predefined standard behavior template one by one, analyze the spatial position changes and time differences of each behavior through frame-by-frame analysis, record and classify the change amounts, set different thresholds for spatial position offset and time offset respectively and group according to the offset amounts, count the occurrence frequency and time distribution of the behavior offset, and at the same time draw a time distribution histogram through statistical methods to record the concentrated distribution characteristics of the offset behavior at different time points. In addition, by statistically analyzing the overall data of each type of offset behavior, calculate the average value and distribution range of the offset behavior for the overall evaluation of the offset situation of the operator's behavior, and obtain the specific measurement data of the degree of deviation of the operator's behavior.

[0076] Compare the operator's behavior with the standard behavior frame by frame, identify the behavior deviation, determine the degree of deviation, and use the formula:

[0077]

[0078] Calculate the behavior deviation metric value D, count the occurrence frequency and time distribution of the deviated behavior to obtain the behavior deviation metric, where α k represents the key weight of the kth frame, O k represents the operator's behavior data in the kth frame, S k represents the standard behavior data in the kth frame, and m is the total number of frames;

[0079] Consider that there are three frames of data in one operation cycle, and the following data is collected:

[0080] α1 = 0.3, O1 = 10, S1 = 8;

[0081] α2 = 0.5, O2 = 14, S2 = 10;

[0082] α3 = 0.2, O3 = 7, S3 = 7;

[0083] For each frame, calculate the absolute difference multiplied by the weight:

[0084] α1|O1 - S1| = 0.3×|10 - 8| = 0.6;

[0085] α2|O2 - S2| = 0.5×|14 - 10| = 2.0;

[0086] α3|O3 - S3| = 0.2×|7 - 7| = 0.0;

[0087] Sum them up and calculate the overall deviation measure:

[0088]

[0089] The result shows that the overall behavior deviation measure is 0.87, indicating that there is a certain degree of deviation between the operator's behavior and the standard behavior during the monitoring period, reflecting the overall deviation degree of the behavior, and is used to evaluate the operator's behavior standardization level.

[0090] The normative evaluation sub-module performs behavior compliance checks based on the behavior deviation measure, compares with the safety operation standard, evaluates the compliance range of the behavior deviation, quantifies the potential impact of the deviation on the safety baseline, and obtains the behavior deviation index;

[0091] Compare the deviation data with the safety operation standard, extract the deviation data related to safety compliance, analyze the potential impact of the data on the operation safety baseline item by item, combine with the pre-set scoring table, assign different importance weights to each deviation data, analyze the influence degree of different deviation factors according to the weights, then classify all deviation factors, calculate the compliance range and importance scores of each deviation category, compare the scores with the requirements of the safety standard, quantitatively describe the safety impact of the behavior deviation, and obtain an overall behavior deviation index.

[0092] As Figure 2 and Figure 5 shown, the real-time risk management module includes a behavior pattern recognition sub-module, a safety level classification sub-module, and a warning adjustment sub-module;

[0093] The behavior pattern recognition sub-module identifies behavior patterns that deviate from standard operations based on behavior deviation metrics, classifies them according to the degree of deviation from the standard, determines the behaviors that belong to deviations from standard operations, and obtains a behavior pattern recognition list.

[0094] Extract features of behaviors that deviate from standard operations. Based on the generated deviation data set, identify the specific deviation types and characteristic parameters of each behavior, extract the key information in the deviation parameter set, including the time period when the behavior occurs, location features, deviation nature, and frequency. Perform a classification operation according to the degree of deviation of the deviation parameters from the standard behavior template, divide the degree of deviation into different classification levels, and determine the behavior pattern features corresponding to each type of deviation by sorting the classification levels, and output a behavior pattern recognition list containing specific classification levels.

[0095] The safety level classification sub-module classifies the monitored behavior patterns according to the risk severity, evaluates the potential threat of each behavior pattern to safety, and divides the safety level according to the risk severity to obtain risk level distribution information.

[0096] Extract all the identified behavior pattern feature data, conduct a risk assessment on each behavior pattern according to the classification level. The assessment content includes the threat degree caused by each type of behavior pattern in actual operations. Refer to the risk level parameters defined in historical data and safety operation standards, divide the behavior patterns according to the risk level, and determine the risk level distribution information by statistically analyzing the distribution frequency of each type of behavior pattern in different risk levels, and at the same time organize and output the association information between each behavior pattern and its risk level.

[0097] Classify the monitored behavior patterns according to the risk severity, evaluate the potential threat of each behavior pattern to safety, and use the formula:

[0098]

[0099] And divide the safety level according to the risk severity to obtain risk level distribution information, where R i represents the risk level of the i-th behavior pattern, β j represents the risk coefficient of the j-th behavior feature, f(B j ) represents the risk function value of the j-th behavior feature, and n represents the total number of behavior features.

[0100] There are three behavior features, and the corresponding risk coefficient β j and the risk function value f(B j ) are as follows:

[0101] β1 = 0.5, f(B1) = 20;

[0102] β2 = 1.2, f(B2) = 15;

[0103] β3 = 0.8, f(B3) = 10;

[0104] β j The value is obtained from the relevant security analysis report, where f(B j ) is obtained through the analysis of the correlation between the behavior pattern and known risk events. For example, f(B j ) can be calculated by statistically analyzing the frequency of security events caused by similar past behaviors. The calculation formula is as follows:

[0105] Calculate the product of each item and its square root:

[0106]

[0107]

[0108] Calculate the value of R i :

[0109]

[0110] The result shows that the average risk level of the i-th behavior pattern is 3.41, which reflects the comprehensive risk level after considering all relevant behavior characteristics, indicates the risk level of the behavior pattern, and is further used for the identification of potential threats and the formulation of preventive measures in the monitoring system.

[0111] The early warning adjustment sub-module automatically adjusts the early warning threshold based on the risk level distribution information, and dynamically adjusts the early warning response level according to the changing security risk data to obtain the risk level response information;

[0112] Conduct hierarchical analysis on various risk levels. According to the latest risk distribution table, extract the level parameters and frequency data of high-risk behavior patterns. By comparing the behavior data corresponding to different risk levels, reset the threshold parameters for early warning triggering, and input the new threshold into the early warning system for automatic update. At the same time, according to the dynamically adjusted early warning threshold, set different response strategy levels. By checking the actual operation results of the adjusted early warning response, generate the risk level response information including the dynamically adjusted early warning threshold and the corresponding response strategy levels.

[0113] As Figure 2 and Figure 6 shown, the monitoring process optimization module includes an early warning information distribution sub-module, a monitoring optimization sub-module, and an early warning effect evaluation sub-module;

[0114] The early warning information distribution submodule extracts the early warning content based on the risk level response information, transmits the high-risk level early warning notification to the monitoring center first, determines the sending status of the notification information, records the sending time and receiving feedback information, and obtains the notification transmission record;

[0115] Extract high-risk warning notification data, classify high-risk warning notifications according to the severity priority of the warning content, transmit the classified notifications to the monitoring center, track the sending status of each notification, record the specific sending time of each notification and the feedback information of the recipient, and when receiving feedback information, extract the confirmation status and response time of the feedback, and generate a notification communication record table based on the transmission order. The record contains the risk level, transmission time, reception confirmation status and feedback time of each notification to ensure that all warning notifications have been delivered.

[0116] The monitoring optimization submodule analyzes the response behavior data of on-site personnel based on the notification transmission records, adjusts the direction and focal range of the monitoring camera, monitors the dynamic changes of the on-site operation area in real time, corrects the monitoring blind spot position, and generates monitoring adjustment details;

[0117] Collect and analyze on-site response data in the monitoring area, extract personnel behavior data recorded during the response process, adjust the direction and focal range of the monitoring camera according to dynamic changes in the operation area, and ensure that key positions in the operation area are within the monitoring range by real-time correction of the camera's viewing angle and focal length. Adjust any monitoring blind spots that appear, analyze the areas missed in the monitoring screen, reset the camera's optimal position and viewing angle, and record the adjusted monitoring parameters in the monitoring adjustment log, including the specific camera direction, focal range, viewing angle correction time, etc.

[0118] The early warning effect evaluation submodule screens the received early warning feedback data based on the monitoring adjustment details, counts the response time and coverage of the different early warning levels, compares them with the processing of the early warning information and the standard processing requirements, evaluates the early warning response effect, and obtains the early warning effect evaluation results;

[0119] The received warning feedback data is sorted and screened, the response time data of different warning levels are extracted, and the coverage of each warning level is counted. The actual processing of the warning information is compared with the standard processing requirements. By analyzing the differentiated response time and coverage, the processing efficiency of each warning level is counted, and the overall warning response effect is evaluated based on the quality of the processing efficiency. At the same time, the analysis results are recorded in the warning effect evaluation table, including the response time, coverage, processing efficiency and their comparison results.

[0120] like Figure 2 and Figure 7As shown in the figure, the behavior pattern update module includes a behavior analysis sub-module, a behavior comparison sub-module, and a database update sub-module;

[0121] Based on the early warning effect evaluation results, the behavior analysis sub-module performs real-time analysis on the behavior data captured in the video stream, extracts key behavior features, separates and marks key behavior changes, and obtains the behavior feature extraction results;

[0122] Extract a continuous frame sequence from the video stream. Determine the key frames containing significant behavior changes in the video through inter-frame difference analysis, and mark the key frames. During the marking process, extract the behavior feature data within the frames. The feature data includes the time point of the behavior, the action direction, the amplitude, and the spatial range involved. Separate the data into different sub-categories, and each sub-category independently records the specific behavior features and change information. Subsequently, conduct a secondary screening of the key behavior changes, eliminate the behavior categories that do not meet the feature change conditions, and generate a feature extraction result table through the classified behavior feature data. The result table contains the classification, time point, and action range of the key behaviors.

[0123] Based on the behavior feature extraction results, the behavior comparison sub-module compares the current behavior data with the behavior patterns stored in the database, identifies new or changed behavior trends, and obtains the behavior trend comparison information;

[0124] Compare the extracted current behavior features with the behavior pattern data stored in the database. Gradually match the behavior templates according to the classification and time point of each behavior feature, and extract the feature parameters similar to the current behavior in the database, including the action type, amplitude range, and time distribution pattern. Divide the matching degree between the current behavior and the database template into different matching levels. For the behavior features with low matching degree or not matched, further analyze the new behavior trends they represent. By comparing their frequencies, change amplitudes, and spatial distributions, classify and mark the unmatched behaviors, and organize the change trend information of the behaviors.

[0125] Based on the behavior trend comparison information, the database update sub-module incorporates the newly mined behavior trends and key data points into the database and optimizes the data structure to obtain an updated behavior database;

[0126] Enter the identified new behavior trends and corresponding key data points into the database. During the entry process, number and classify the new behavior trends, insert the behavior trend data into the existing database structure in chronological order, and optimize the database according to the structural characteristics of the new data. The optimization content includes adjusting the storage method of data indexes, adding index tags to frequently accessed data, cleaning duplicate or invalid data. After the optimization is completed, re-retrieve and verify all the newly added behavior trends and key data to ensure data consistency and integrity, and finally obtain an updated behavior database.

[0127] As Figure 2 and Figure 8 shown, the emergency response correction module includes a behavior guidance sub-module and an accident management sub-module;

[0128] Based on the updated behavior database, the behavior guidance sub-module provides behavior guidance to operators, analyzes unsafe behavior patterns, and through simulation training and behavior correction, refines the understanding results of operators on risk behaviors to obtain behavior correction guidance data;

[0129] First, extract the dataset on unsafe behavior patterns in the database, analyze the classification information of unsafe behaviors and the corresponding key feature parameters, including behavior deviations, action frequencies, and spatial position ranges in the operation steps. Use the parameters to construct simulation training tasks, input the feature data of unsafe behaviors into the simulation training module to generate corresponding scenarios, guide operators to identify risk behaviors in the simulation scenarios, record the feedback time of operators on risk behaviors and the behavior adjustment process, and at the same time decompose each operation step. By recording and analyzing the response characteristics of personnel during behavior correction, including operation accuracy and adjustment frequency, finally integrate all simulation training results and correction process data to form complete behavior correction guidance data.

[0130] Based on the behavior correction guidance data, the accident management sub-module formulates emergency shutdown and intervention measures for high-risk behaviors, adjusts the operation process and matches potential safety threats, optimizes the accident occurrence probability and intervention effectiveness to obtain an accident prevention and control management log;

[0131] Extract the classification information involving high-risk behaviors, formulate an emergency shutdown plan for each type of high-risk behavior according to the classification information. The emergency shutdown plan includes the setting of shutdown condition parameters, trigger steps, and recovery processes after shutdown. Match the shutdown parameters with the actual operation process, identify the steps in the operation process associated with high-risk behaviors, correspond the potential safety threat points to the process, adjust the operation process to reduce the triggering possibility of high-risk behaviors, and at the same time analyze the operation efficiency of the adjusted process. Combine the adjustment results to generate an intervention effectiveness evaluation record, and integrate all emergency shutdown plans and process optimization data into the accident management log to form an accident prevention and control management log.

[0132] It should be understood that the term "and / or" in this article is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context.

[0133] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or a similar expression means any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or plural.

[0134] It should be understood that in various embodiments of the present invention, the magnitude of the sequence numbers of the above - mentioned processes does not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0135] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0136] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the devices, apparatuses, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0137] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0138] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0139] In addition, in each embodiment of the present invention, each functional unit may be integrated into one processing unit, may exist physically alone for each unit, or two or more units may be integrated into one unit.

[0140] If the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0141] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A chemical production safety intelligent management system based on video monitoring, characterized in that: The system comprises: The video stream optimization module adjusts the image resolution and frame rate based on the real-time video stream of chemical production captured by the camera, eliminates the influence of background noise and lighting changes, decodes and compresses the signal, and corrects color distortion and motion blur to obtain optimized video stream quality. The behavior analysis and evaluation module tracks the behavior changes between key frames based on the optimized video stream quality, identifies deviations from standard operations, and evaluates the standardization of the operation behavior based on the deviations to obtain a behavior deviation index; The real-time risk management module identifies the behavior patterns that deviate from the standard operation based on the behavior deviation indicators, automatically classifies the safety level, calculates and predicts the probability of occurrence of multi-level risks, and automatically adjusts the warning threshold to obtain risk level response information; The monitoring process optimization module distributes the warning notification to the monitoring center and the on-site operation area based on the risk level response information, monitors the personnel status in the on-site operation area, adjusts the monitoring parameters according to the personnel feedback information, evaluates the warning response effect, and obtains the warning effect evaluation result; The behavior pattern update module continuously monitors the behavior pattern characteristics in the video stream based on the warning effect evaluation result, compares them with the historical behavior data, identifies new behavior trends or behavior changes, and obtains an updated behavior database; The emergency response correction module provides behavioral correction guidance to operators based on the updated behavior database, formulates emergency shutdown and intervention measures for the monitored risk behaviors, and obtains an accident prevention and control management log.

2. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The optimized video stream quality includes signal integration, dynamic range, and frame integration; the behavioral deviation indicators include deviation intensity, consistency index, and behavioral deviation frequency; the risk level response information includes risk discrimination threshold, response level classification, and warning signal strength; the warning effect evaluation results include response effectiveness, reaction adjustment time, and monitoring efficiency; the updated behavior database specifically includes behavior pattern recognition results, pattern change records, and behavior sequence analysis results.

3. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The video stream optimization module includes an image quality adjustment submodule and an image stabilization submodule; The image quality adjustment submodule adjusts the image resolution and frame rate based on the real-time video stream of chemical production captured by the camera to match the environmental requirements of differential monitoring and optimize the impact of background noise and illumination changes to obtain video quality optimized records; The signal processing submodule decodes and compresses the signal of the video quality optimization record, adjusts the transmission parameters and storage requirements of the data, corrects color distortion and motion blur, optimizes the clarity of the monitoring picture, and obtains the optimized video stream quality.

4. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The behavior analysis and evaluation module includes a behavior tracking submodule, a deviation analysis submodule, and a normative evaluation submodule; Based on the optimized video stream quality, the behavior tracking submodule monitors the behavior in the video frame sequence, captures the behavior changes between key frames, marks each key frame with a timestamp, associates the behavior event with the operation area, refines the corresponding relationship between time and space data, and obtains behavior tracking mapping data; The deviation analysis submodule compares the operator's behavior with the standard behavior frame by frame based on the behavior tracking mapping data, identifies the behavior deviation, determines the degree of deviation, and counts the frequency and time distribution of the deviation behavior to obtain the behavior deviation metric; The normative assessment submodule performs a behavioral compliance check based on the behavioral deviation metric, compares it with the safety operation standard, evaluates the compliance scope of the behavioral deviation, quantifies the potential impact of the deviation on the safety baseline, and obtains the behavioral deviation index.

5. The chemical production safety intelligent management system based on video monitoring according to claim 4 is characterized in that: The operator's behavior is compared frame by frame with the standard behavior to identify the behavior deviation and determine the degree of deviation using the formula: Calculate the behavior deviation metric D, count the frequency and time distribution of the deviation behavior, and obtain the behavior deviation metric, where α k represents the critical weight of the kth frame, O k represents the operator behavior data in the kth frame, S k represents the standard behavior data in the kth frame, and m is the total number of frames.

6. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The real-time risk management module includes a behavior pattern recognition submodule, a security level classification submodule, and an early warning adjustment submodule; The behavior pattern recognition submodule recognizes the behavior pattern that deviates from the standard operation based on the behavior deviation index, classifies the behavior according to the degree of deviation between the behavior and the standard, determines the behavior that deviates from the standard operation, and obtains a behavior pattern recognition list; The security level classification submodule classifies the monitored behavior patterns according to the risk severity based on the behavior pattern recognition list, evaluates the potential threat to security of each behavior pattern, and divides the security level according to the risk severity to obtain risk level distribution information; The early warning adjustment submodule automatically adjusts the early warning threshold based on the risk level distribution information, and dynamically adjusts the early warning response level according to the changing security risk data to obtain risk level response information.

7. The chemical production safety intelligent management system based on video monitoring according to claim 6 is characterized in that: The monitored behavior patterns are classified according to the severity of the risk, and the potential threat of each behavior pattern to security is evaluated using the formula: And divide the security level according to the severity of the risk to obtain the risk level distribution information, where R i represents the risk level of the i-th behavior pattern, β j represents the risk factor of the jth behavior feature, f(B j ) represents the risk function value of the jth behavior feature, and n represents the total number of behavior features.

8. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The monitoring process optimization module includes a warning information distribution submodule, a monitoring optimization submodule, and a warning effect evaluation submodule; The early warning information distribution submodule extracts the early warning content based on the risk level response information, transmits the early warning notification of the high risk level to the monitoring center first, determines the sending status of the notification information, records the sending time and the received feedback information, and obtains the notification transmission record; The monitoring optimization submodule analyzes the response behavior data of the on-site personnel based on the notification communication record, adjusts the direction and focal range of the monitoring camera, monitors the dynamic changes of the on-site operation area in real time, corrects the monitoring blind spot position, and generates monitoring adjustment details; The early warning effect evaluation submodule screens the received early warning feedback data based on the monitoring adjustment details, counts the response time and coverage of the difference warning levels, compares them with the processing of the early warning information and the standard processing requirements, evaluates the early warning response effect, and obtains the early warning effect evaluation result.

9. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The behavior pattern update module includes a behavior analysis submodule, a behavior comparison submodule, and a database update submodule; The behavior analysis submodule performs real-time analysis on the behavior data captured in the video stream based on the warning effect evaluation result, extracts key behavior features, separates and marks key behavior changes, and obtains behavior feature extraction results; The behavior comparison submodule compares the current behavior data with the behavior patterns stored in the database based on the behavior feature extraction results, identifies new or changed behavior trends, and obtains behavior trend comparison information; The database updating submodule incorporates the newly mined behavior trends and key data points into the database based on the behavior trend comparison information, and optimizes the data structure to obtain an updated behavior database.

10. The chemical production safety intelligent management system based on video monitoring according to claim 1 is characterized in that: The emergency response correction module includes a behavior guidance submodule and an accident management submodule; The behavior guidance submodule provides behavior guidance to the operator based on the updated behavior database, analyzes unsafe behavior patterns, and refines the operator's understanding of risky behaviors through simulation training and behavior correction to obtain behavior correction guidance data; Based on the behavior correction guidance data, the accident management submodule formulates emergency shutdown and intervention measures for high-risk behaviors, adjusts operating procedures and matches potential safety threats, optimizes the probability of accidents and the effectiveness of intervention, and obtains an accident prevention and control management log.

Images