Management system and method of intelligent management cabinet based on communication machine room

Abstract

The invention relates to the field of communication machine room management cabinets, and particularly discloses a management system and method for an intelligent management cabinet based on a communication machine room, and the method comprises the steps: firstly obtaining an internal monitoring video, collected by a camera, of the communication machine room management cabinet and smoke values, collected by a smoke sensor, of the environment of the communication machine room management cabinet at a plurality of preset time points; and then feature extraction and correlation analysis are carried out on the two by using a deep learning technology, and finally a classification result is obtained through a classifier to judge whether a communication machine room fire safety early warning is sent out, so that the workload of manual monitoring is reduced, the intelligent level and the operation efficiency of the system are improved, and the overall safety of the communication machine room is improved.

Description

Technical Field

[0001] The present application relates to the field of management cabinets in communication rooms, and more specifically, to a management system and method for an intelligent management cabinet based on a communication room. Background Art

[0002] With the continuous development of communication technology, the number of communication systems and their supporting equipment is increasing day by day, and the communication room has become an important part of the telecommunication network. In the field of communication room technology, a line management cabinet is installed to control the internal circuit. There are many lines inside the line management cabinet, and there are many accessories and drive loads. If any circuit or electrical equipment fails in any link, it will lead to the occurrence of fire accidents. The communication room is a major hidden danger of electrical fire.

[0003] At present, the management of many communication rooms adopts 24-hour dedicated personnel to regularly inspect the room line management cabinet. This not only increases the burden on management personnel, but also often fails to eliminate faults in time, and the early warning method is relatively lagging, which is extremely detrimental to the safe operation of the room.

[0004] Therefore, a management system and method for an intelligent management cabinet based on a communication room is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a management system and method based on an intelligent management cabinet of a communication room, which first obtains the internal monitoring video of the communication room management cabinet collected by a camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by a smoke sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the classification result through a classifier to determine whether to issue a fire safety warning for the communication room, thereby reducing the workload of manual monitoring, thereby improving the intelligence level and operational efficiency of the system, and improving the overall safety of the communication room.

[0006] According to one aspect of the present application, a management system for an intelligent management cabinet based on a communication room is provided, which includes:

[0007] The communication room management cabinet data acquisition module is used to obtain the internal monitoring video of the communication room management cabinet collected by the camera and the smoke value of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor;

[0008] A communication room management cabinet data extraction module is used to extract the communication room management cabinet internal monitoring focus feature vector and the communication room management cabinet smoke time series correlation feature vector from the communication room management cabinet internal monitoring video collected by the camera and the communication room management cabinet environment smoke values ​​at multiple predetermined time points collected by the smoke sensor;

[0009] The fire safety warning judgment module is used to determine whether to issue a fire safety warning for the communication room based on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet.

[0010] According to another aspect of the present application, a management method for an intelligent management cabinet based on a communication room is provided, comprising:

[0011] Obtaining the internal monitoring video of the management cabinet of the communication room collected by the camera and the smoke values ​​of the management cabinet environment of the communication room collected by the smoke sensor at multiple predetermined time points;

[0012] Extracting the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet from the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor;

[0013] Based on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series association feature vector of the communication room management cabinet, it is determined whether to issue a fire safety warning for the communication room.

[0014] Compared with the prior art, the present application provides a management system and method for an intelligent management cabinet based on a communication room, which first obtains the internal monitoring video of the communication room management cabinet collected by a camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by a smoke sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the classification result through a classifier to determine whether to issue a fire safety warning for the communication room, thereby reducing the workload of manual monitoring, thereby improving the intelligence level and operational efficiency of the system, and improving the overall safety of the communication room. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a block diagram of a management system based on an intelligent management cabinet in a communication room according to an embodiment of the present application.

[0017] Figure 2 It is a block diagram of a data extraction module of a communication room management cabinet in a management system based on an intelligent management cabinet of a communication room according to an embodiment of the present application.

[0018] Figure 3 It is a block diagram of the internal monitoring feature extraction unit of the management cabinet in the management system of the intelligent management cabinet based on the communication room according to the embodiment of the present application.

[0019] Figure 4 It is a block diagram of a fire safety early warning judgment module in a management system of an intelligent management cabinet based on a communication room according to an embodiment of the present application.

[0020] Figure 5 The present invention is a flowchart of a management method of an intelligent management cabinet based on a communication room according to an embodiment of the present application.

[0021] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] Figure 1 FIG. 1 is a block diagram of a management system based on an intelligent management cabinet in a communication room according to an embodiment of the present application. Figure 1 As shown, according to the management system 100 of the intelligent management cabinet based on the communication room according to the embodiment of the present application, it includes: a communication room management cabinet data acquisition module 110, which is used to obtain the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor; a communication room management cabinet data extraction module 120, which is used to extract the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet from the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor; a fire safety warning judgment module 130, which is used to judge whether to issue a fire safety warning for the communication room based on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet.

[0024] In the management system 100 based on the intelligent management cabinet of the communication room, the communication room management cabinet data acquisition module 110 is used to obtain the internal monitoring video of the communication room management cabinet collected by the camera and the smoke value of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor. It should be understood that with the rapid advancement of communication technology, the number of communication systems and related equipment has continued to increase, making the communication room a vital part of the telecommunications network. In order to manage the circuits in these rooms, line management cabinets are usually installed. However, due to the complex internal circuits and numerous components of these cabinets, once there is a problem with the circuit or electrical equipment, it is possible to cause a fire. Therefore, there is a greater risk of electrical fire in the communication room. At present, many communication rooms rely on 24-hour special personnel on duty for regular inspections, which not only increases the workload of managers, but also may not be able to quickly discover and solve faults, resulting in delayed early warning responses, thereby affecting the safe operation of the room. Therefore, in the technical solution of the present application, by obtaining the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor, and combining with deep learning technology, it is determined whether it is necessary to issue a fire safety warning for the communication room, so as to reduce the burden of manual monitoring, thereby improving the intelligence and operational efficiency of the system, and enhancing the overall safety of the communication room.

[0025] Specifically, the purpose of obtaining the internal surveillance video of the management cabinet of the communication room collected by the camera and the environmental smoke value collected by the smoke sensor is to realize a comprehensive security monitoring and early warning system. The internal surveillance video captured by the camera can provide real-time visual information to help identify abnormal conditions inside the room, such as equipment failure, personnel intrusion or other potential risks. On the other hand, the smoke value data provided by the smoke sensor is the key to detecting fire or smoke leakage. By recording the smoke concentration at multiple predetermined time points, the system can establish a time series of smoke changes and identify possible signs of fire or trends of smoke leakage. The time series analysis of smoke values ​​is crucial for the early warning system because sudden changes in smoke concentration are often early signs of fire. By comprehensively analyzing the camera video and smoke sensor data, the system can achieve more efficient security monitoring, timely detect and respond to possible risks, and ensure the safe operation of the communication room. This comprehensive monitoring method not only improves the reliability of early warning, but also provides more sufficient data support for the formulation of emergency response measures.

[0026] In the management system 100 based on the intelligent management cabinet of the communication room, the communication room management cabinet data extraction module 120 is used to extract the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet from the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor. In this way, the system can combine video monitoring and environmental data for more comprehensive analysis. This multi-dimensional data fusion not only improves the accuracy and response speed of monitoring, but also enhances the sensitivity and reliability of the early warning system to potential security threats in complex environments.

[0027] Figure 2 FIG. 1 is a block diagram of a data extraction module for a communication room management cabinet in a management system based on an intelligent management cabinet of a communication room according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the communication room management cabinet data extraction module 120 includes: a management cabinet internal monitoring feature extraction unit 121, which is used to perform feature extraction on the internal monitoring video of the communication room management cabinet collected by the camera to obtain the communication room management cabinet internal monitoring focus feature vector; a management cabinet environment smoke feature extraction unit 122, which is used to perform feature extraction on the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor to obtain the communication room management cabinet smoke time series correlation feature vector.

[0028] It should be understood that feature extraction converts raw video data into analyzable high-level information, allowing the system to extract meaningful features from a large amount of image data. These feature vectors can represent key details and dynamic changes in the video, such as equipment operating status, personnel location, abnormal behavior, etc. Through this conversion, the system can effectively filter out redundant information and noise and focus on the parts that are critical to the safety of the computer room. By applying feature extraction to video data, the system can realize intelligent monitoring and analysis functions. This intelligent analysis capability can significantly improve the accuracy and response speed of early warnings and reduce errors and delays in human monitoring.

[0029] Furthermore, feature extraction is performed on the smoke values ​​of the communication room management cabinet environment collected by the smoke sensor at multiple predetermined time points to obtain the smoke time series correlation feature vector in order to achieve accurate fire warning and environmental monitoring. The smoke value data provided by the smoke sensor is usually continuous, but directly using these raw data for analysis may not be able to effectively capture potential fire risks and abnormal trends. By extracting features and converting these data into time series correlation feature vectors, deeper insights and analysis capabilities can be provided.

[0030] Figure 3FIG. 1 is a block diagram of a monitoring feature extraction unit for an internal management cabinet in a management system based on an intelligent management cabinet in a communication room according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the management cabinet internal monitoring feature extraction unit 121 includes: a management cabinet internal monitoring data preprocessing subunit 1211, which is used to perform data preprocessing on the internal monitoring video of the communication room management cabinet collected by the camera to obtain multiple communication room management cabinet internal monitoring enhanced key frames; a management cabinet internal monitoring feature encoding subunit 1212, which is used to perform feature encoding on the multiple communication room management cabinet internal monitoring enhanced key frames to obtain the communication room management cabinet internal monitoring focus feature vector.

[0031] It should be understood that the video recorded by the camera may be affected by light changes, motion blur or other environmental factors, which will reduce the clarity and details of the video. Through data preprocessing technology, these adverse effects can be reduced, the clarity can be improved, and the key details in the video can be more obvious.

[0032] Furthermore, feature encoding is performed on multiple enhanced key frames of internal monitoring of the management cabinet of the communication room to obtain the feature vector of interest for internal monitoring of the management cabinet of the communication room, in order to achieve efficient and accurate security monitoring and event analysis. Feature encoding converts visual information into structured data, so that the computer system can effectively process and analyze the video content. The original enhanced key frame contains a large amount of pixel data, and directly processing this data may result in excessive computational burden. Through feature encoding, the system extracts and compresses important information from the image into a feature vector with high dimension. This conversion makes subsequent processing more efficient. By encoding the key frame into a feature vector, the system can uniformly compare and analyze the monitoring data at different times and perspectives. This standardized form of feature vectors enables the system to easily integrate monitoring data from different cameras and identify long-term trends and abnormal patterns. For large-scale monitoring systems, this feature encoding method can effectively support data analysis across devices and time periods.

[0033] In a specific embodiment of the present application, the management cabinet internal monitoring data preprocessing subunit 1211 includes: extracting multiple communication room management cabinet internal monitoring key frames from the communication room management cabinet internal monitoring video collected by the camera; passing the multiple communication room management cabinet internal monitoring key frames through the communication room management cabinet internal monitoring convolutional neural network as a noise reducer to obtain the multiple communication room management cabinet internal monitoring enhanced key frames.

[0034] It should be understood that extracting multiple key frames helps capture and record important moments in the video. The environment inside the management cabinet of the communication room may undergo various changes, such as equipment operation, personnel activities, or sudden abnormal events. A single frame may not fully reflect the entire scene, and multiple key frames can provide more perspectives and information at moments, thereby ensuring that important events and states are fully recorded. This approach enables the monitoring system to analyze and respond to key events in the environment more accurately. Extracting key frames can significantly reduce the complexity and computational burden of data processing. Raw video data usually contains a large number of consecutive frames, many of which may contain similar information, resulting in a waste of computing and storage resources. By extracting representative key frames, the system can reduce the amount of data that needs to be processed while retaining key scene changes and important information. This approach not only improves processing efficiency, but also reduces the cost of data storage and management.

[0035] Furthermore, the multiple communication room management cabinet internal monitoring key frames are processed by a convolutional neural network (CNN) as a noise reducer to improve the clarity and quality of the video image and ensure that the monitoring system can detect and analyze key events more effectively. The convolutional neural network has a powerful image processing capability and can significantly reduce the noise in the image. The image in the monitoring video may be noisy or blurred due to insufficient light, camera quality or other environmental factors. The convolutional neural network can automatically learn and extract important features in the image through its multi-layer convolution and pooling operations, while suppressing noise and unnecessary interference. Specifically, the multiple communication room management cabinet internal monitoring key frames are input into the encoder of the communication room management cabinet internal monitoring convolutional neural network as a noise reducer, wherein the encoder uses a convolutional layer to perform explicit spatial encoding on the multiple communication room management cabinet internal monitoring key frames to obtain image features; and the image features are input into the decoder of the communication room management cabinet internal monitoring convolutional neural network as a noise reducer, wherein the decoder uses a deconvolution layer to perform deconvolution processing on the image features to obtain the multiple communication room management cabinet internal monitoring enhanced key frames after denoising.

[0036] In a specific embodiment of the present application, the management cabinet internal monitoring feature encoding subunit 1212 includes: passing the multiple communication room management cabinet internal monitoring enhanced key frames through a communication room management cabinet internal monitoring interest region feature encoder based on a target detection network to obtain multiple communication room management cabinet internal monitoring interest regions; arranging the multiple communication room management cabinet internal monitoring interest regions along the sample dimension to obtain a communication room management cabinet internal monitoring interest input tensor; passing the communication room management cabinet internal monitoring interest input tensor through a communication room management cabinet internal monitoring interest feature encoder based on a three-dimensional convolutional neural network to obtain the communication room management cabinet internal monitoring attention feature vector.

[0037] It should be understood that the feature encoder based on the target detection network can accurately identify and extract the regions of interest in the surveillance image. The surveillance keyframe inside the communication room may contain a lot of background information and noise, and some of these areas may not be important for security monitoring and equipment management. By applying the target detection network, the system can automatically identify areas of critical significance, such as equipment operation panels, personnel activity areas, or potential abnormal behavior areas. The extraction of such areas can help the system focus on those parts that are most important to the monitoring task and avoid ineffective processing of irrelevant information. The extraction of regions of interest helps to improve data processing efficiency. The regions of interest in the surveillance video are usually the focus of attention, such as equipment status, personnel behavior, or environmental changes. By extracting these areas from the keyframes, the system can focus only on these important parts during processing and analysis, thereby significantly reducing the amount of calculation and storage requirements. By feature encoding important areas in the image, the system can obtain detailed regional information, such as the precise location, size, and characteristics of the target. This information is crucial for subsequent intelligent analysis, event backtracking, and behavior recognition. Specifically, the communication room management cabinet internal monitoring interest region feature encoder based on the target detection network is used to process the multiple communication room management cabinet internal monitoring enhanced key frames with the following target detection formula to obtain the multiple communication room management cabinet internal monitoring interest regions;

[0038] Among them, the target detection formula is:

[0039]

[0040] in, Enhance keyframes for monitoring inside management cabinets of multiple communication rooms. is the anchor frame, Monitor areas of interest inside management cabinets for multiple communication rooms, Indicates classification, Indicates regression.

[0041] Furthermore, in monitoring systems, regions of interest usually contain important visual information, such as equipment status, personnel activities, etc. By arranging these regions according to the sample dimension, the data of each region can be integrated into a unified input tensor. This structured data format enables the system to process multiple samples in a consistent manner, simplifies data management and operation, and improves the efficiency of analysis.

[0042] Furthermore, the input tensor of interest for monitoring the internal management cabinet of the communication room is processed through a feature encoder based on a three-dimensional convolutional neural network (3D CNN) to obtain the feature vector of interest for monitoring the internal management cabinet of the communication room, in order to improve the understanding of monitoring data, extract key features, and optimize the overall performance of the monitoring system. Three-dimensional convolutional neural networks (3D CNN) are particularly suitable for processing data with time and space dimensions. The monitoring data inside the communication room usually contains information of time series and spatial distribution, such as dynamic changes and spatial layout in video frames. 3D CNN can process information of these dimensions at the same time and capture the spatiotemporal characteristics of the target in the video, thereby understanding and analyzing the monitoring data more comprehensively than the traditional two-dimensional convolutional neural network (2DCNN). By using 3D CNN, it is possible to better identify and extract features that are significant in both time series and spatial dimensions, which is crucial for accurately analyzing the monitoring content. Specifically, each layer of the communication room management cabinet internal monitoring feature encoder based on the three-dimensional convolutional neural network performs convolution processing, mean pooling processing based on the local feature matrix and nonlinear activation processing on the input data in the forward pass of the layer, so that the last layer of the communication room management cabinet internal monitoring feature encoder based on the three-dimensional convolutional neural network outputs the communication room management cabinet internal monitoring feature vector, wherein the input of the communication room management cabinet internal monitoring feature encoder based on the three-dimensional convolutional neural network is the communication room management cabinet internal monitoring input tensor of interest.

[0043] In a specific embodiment of the present application, the management cabinet environment smoke feature extraction unit 122 includes: arranging the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor into multiple management cabinet smoke input vectors; passing the multiple management cabinet smoke input vectors through the communication room management cabinet smoke timing encoder to obtain the communication room management cabinet smoke timing associated feature vector.

[0044] It should be understood that time series data can reflect the trend of smoke concentration over time, helping the system to identify long-term change patterns, periodic fluctuations, and sudden abnormal situations. By integrating smoke values ​​at different time points into an input vector, the system can more comprehensively understand the trajectory of smoke concentration changes, thereby providing more accurate risk assessment and warning information. This arrangement provides structured input for data analysis and model training.

[0045] Furthermore, the smoke input vectors of multiple management cabinets are processed by the smoke time series encoder of the management cabinet in the communication room in order to deeply explore the time series relationship in the time series data and optimize the abnormal detection and prediction capabilities of the monitoring system. The smoke time series encoder can capture the time series features in the time series data. The smoke input vectors at multiple time points reflect the changing trend of smoke concentration over time. Through the processing of the time series encoder, the dynamic changes and historical dependencies in these time series data can be effectively extracted. By learning the long-term dependencies in time, the time series encoder can identify potential trends, periodic changes and emergencies, which is crucial for the real-time response of the monitoring system. The smoke time series encoder can extract complex patterns and associated features in the time series data. The smoke concentration of the management cabinet is affected not only by the current state, but also by the past state. By encoding the data at multiple time points, the time series encoder can transform these complex time series patterns into highly abstract feature vectors. These feature vectors not only contain the changing trend of smoke concentration, but also capture the correlation between different time points, thereby improving the system's ability to identify abnormal situations. Specifically, the multiple management cabinet smoke input vectors are fully-connected using the fully-connected layer of the communication room management cabinet smoke timing encoder to extract high-dimensional implicit features of the eigenvalues ​​at each position in the multiple management cabinet smoke input vectors; and, the multiple management cabinet smoke input vectors are one-dimensionally encoded using the one-dimensional convolutional layer of the communication room management cabinet smoke timing encoder to extract high-dimensional implicit correlation features of the correlation between the eigenvalues ​​at each position in the multiple management cabinet smoke input vectors.

[0046] In the management system 100 based on the intelligent management cabinet of the communication room, the fire safety warning judgment module 130 is used to determine whether to issue a fire safety warning for the communication room based on the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet. It should be understood that the internal monitoring feature vector usually includes parameters related to the health of the equipment, such as the operating status, temperature, humidity, etc. of the equipment. These data can help the monitoring system understand the environment and equipment conditions inside the room, such as whether the equipment is overheated, faulty or working abnormally. Overheating or failure of the equipment may often cause a fire, and these internal monitoring feature vectors provide real-time feedback on the health status of the equipment to help identify potential fire risks. The smoke time series correlation feature vector provides time series data of smoke concentration, which can reveal the trend of smoke concentration changes and its relationship with time. The feature vector extracted by the time series encoder can capture the short-term and long-term change patterns of smoke concentration, including sudden concentration surges and slow trend changes. Combining and comprehensively analyzing these two types of data can effectively identify real fire risks and issue warnings in time to prevent potential fires from causing serious damage to the communication room.

[0047] Figure 4 FIG. 1 is a block diagram of a fire safety early warning judgment module in a management system based on an intelligent management cabinet in a communication room according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the fire safety warning judgment module 130 includes: a communication room management cabinet security feature fusion unit 131, which is used to perform perspective conversion matching based on topological correlation correction on the internal monitoring focus feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet to obtain a fire safety warning judgment classification feature vector; a communication room fire safety warning classification judgment unit 132, which is used to pass the fire safety warning judgment classification feature vector through a classifier to obtain a classification result, and the classification result is used to determine whether to issue a fire safety warning for the communication room.

[0048] It should be understood that fusing these two types of feature vectors can more comprehensively assess fire safety risks. The internal monitoring feature vector provides immediate feedback on the equipment status and environmental conditions, while the smoke time series feature vector supplements the temporal dynamic information about the occurrence of fire. By fusing these two types of data, the impact of equipment abnormalities and changes in smoke concentration can be comprehensively considered, thereby improving the ability to identify fire risks. For example, when the temperature of the equipment rises abnormally and the smoke concentration begins to increase significantly, this multi-dimensional data fusion can more accurately determine whether there is a fire hazard.

[0049] In particular, in the technical solution of the present application, it is considered that the internal monitoring feature vector of the communication room management cabinet is extracted from the internal monitoring video of the communication room management cabinet collected by the camera, and the main focus is on the visual information and key areas in the video. The smoke time series correlation feature vector of the communication room management cabinet is extracted from the smoke values ​​of multiple predetermined time points collected by the smoke sensor, and the main focus is on the time series changes of the smoke values. Since the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet are extracted from different data sources and processing paths, they represent different aspects of the fire safety monitoring of the communication room. Due to the essential differences of these features and the different extraction methods, these two feature vectors have significant feature heterogeneity. This heterogeneity may cause the following problems in the feature fusion process: First, the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet may exist on different manifolds, which means that they have different distribution laws in their respective spaces. Directly splicing them together may result in the final feature vector not having a clear manifold in the new space, that is, the feature manifold smoothness is insufficient. Second, since the two feature vectors come from different data sources and processing paths, they may not be well mapped to each other in the fused feature space, that is, one type of feature may not accurately reflect the feature changes of another type, resulting in increased uncertainty in feature expression. In order to solve the above problems, in the technical solution of the present application, the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet are matched based on topological correlation correction.

[0050] Specifically, in an embodiment of the present application, a perspective conversion matching based on topological correlation correction is performed on the internal monitoring concern feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet to obtain a fire safety warning judgment classification feature vector, including: constructing a communication room management cabinet internal monitoring-smoke timing distance topological matrix between the internal monitoring concern feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet; constructing a communication room management cabinet internal monitoring-smoke timing feature granularity interaction matrix between the internal monitoring concern feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet; based on the communication room management cabinet internal monitoring-smoke timing distance topological matrix, the communication room The internal monitoring-smoke timing feature granularity interaction matrix of the management cabinet is topologically corrected to obtain the internal monitoring-smoke timing topological correction association matrix of the communication room management cabinet; the internal monitoring-smoke timing topological correction association matrix of the communication room management cabinet is used as the perspective co-projection space, and the internal monitoring focus feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet are projected to the perspective co-projection space to obtain the perspective conversion internal monitoring focus feature vector of the communication room management cabinet and the perspective conversion smoke timing association feature vector of the communication room management cabinet; the perspective conversion internal monitoring focus feature vector of the communication room management cabinet and the perspective conversion smoke timing association feature vector of the communication room management cabinet are fused to obtain the fire safety early warning judgment classification feature vector.

[0051] In an embodiment of the present application, another implementation expression of performing perspective conversion matching based on topological correlation correction on the internal monitoring feature vector of the communication room management cabinet and the smoke time series associated feature vector of the communication room management cabinet to obtain a fire safety early warning judgment classification feature vector can be:

[0052]

[0053]

[0054]

[0055]

[0056]

[0057]

[0058] in, represents the monitoring focus feature vector inside the management cabinet of the communication room, represents the smoke time series correlation feature vector of the communication room management cabinet, is the distance topological matrix between the internal monitoring attention feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet, that is, , is the distance topology matrix The eigenvalues ​​of the positions, represents the calculation of the first feature vector of interest for monitoring inside the management cabinet of the communication room The characteristic value of the position and the first characteristic vector of the smoke time series correlation of the communication room management cabinet The distance of the eigenvalue of the position, and and are column vectors, represents the distance topology matrix, represents the transpose of a vector, represents matrix multiplication, represents the feature granularity interaction matrix, represents the natural exponential function, represents the topologically corrected incidence matrix, Indicates the perspective conversion of the internal monitoring feature vector of the communication room management cabinet, Indicates the time series correlation feature vector of smoke in the management cabinet of the communication room when the perspective is changed, and represents the weighted hyperparameter, Represents the fire safety early warning judgment classification feature vector.

[0059] That is, in the technical solution of the present application, in order to optimize the correlation analysis between the internal monitoring of the communication room management cabinet and the smoke detection, in order to solve the problem that the internal monitoring feature vector of the communication room management cabinet and the smoke timing correlation feature vector of the communication room management cabinet have relatively significant feature heterogeneity, which leads to the feature manifold and feature space mismatch caused by heterogeneity when the feature fusion of the internal monitoring feature vector of the communication room management cabinet and the smoke timing correlation feature vector of the communication room management cabinet is performed, which will cause the fused feature vector to fail to have the feature manifold smoothness, thereby affecting the certainty of its feature expression. The present application adopts a perspective conversion matching method based on topological correlation correction. This process first involves constructing a distance topological matrix that reflects the relationship between the internal monitoring feature vector of the communication room management cabinet and the smoke timing correlation feature vector of the communication room management cabinet. By treating each eigenvalue in these feature vectors as a node in the graph and using an appropriate distance metric function to define the strength of the connection between the nodes, the present application can create a matrix that can accurately describe the complex topological relationship between the two. This matrix not only reveals the distribution pattern of data points in multidimensional space, but also provides important structural information for subsequent feature analysis.

[0060] Next, this application further explores the granularity-level associations between different attributes within the feature vector, and by calculating the transposed product of the internal monitoring focus feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet, a matrix is ​​formed that can reflect the fine degree of association between the eigenvalues ​​of the two. Subsequently, this granular association matrix was corrected using the previously constructed distance topology matrix, aiming to capture the dynamic changes of the eigenvalues ​​in the topological structure, thereby generating a topologically corrected association matrix that can reflect the deep connection between monitoring and smoke timing. This process is equivalent to allowing the eigenvalues ​​to "wander" on the topological map, so that the original high-dimensional data can be effectively mapped to a low-dimensional space with continuous high-dimensional regression characteristics while maintaining key information.

[0061] After obtaining the corrected correlation matrix, the present application projects the communication room management cabinet internal monitoring attention feature vector and the communication room management cabinet smoke time series correlation feature vector into the new perspective co-projection space, generating two new feature vectors after perspective transformation. The core of this transformation is to find an implicit space that can simultaneously maintain the main features and relative position relationship of the original data, so that the data is reduced in dimension without losing too much useful information.

[0062] Finally, the two perspective-converted feature vectors are integrated into a fire safety warning judgment classification feature vector. This fusion process can select different methods according to actual needs, such as simple weighted average or more advanced machine learning algorithms, with the goal of generating a comprehensive feature vector that includes both monitoring data and smoke detection information. Such a feature vector not only helps to improve the predictive performance of the machine learning model, but also provides strong support for processing cross-modal or multi-source data, greatly improving the understanding and application efficiency of complex data environments.

[0063] Furthermore, the fire safety warning judgment classification feature vector is obtained by fusing the internal monitoring data and the smoke time series data. These feature vectors integrate the dynamic change information of the equipment status, environmental conditions and smoke concentration, and provide a comprehensive description of the fire risk. However, these feature vectors themselves are multidimensional and complex, and it may be challenging to directly interpret these data and make warning decisions. Therefore, it is necessary to use a classifier to process and analyze these data. In the fire safety warning scenario, the classifier maps the fire safety warning judgment classification feature vector to a predefined category, such as "normal" or "warning", and outputs the classification result. This classification result provides a concise decision-making basis, telling the system whether a fire safety warning needs to be issued. Using the classification results to determine whether to issue a fire safety warning in the communication room can effectively transform complex data analysis results into practical actions.

[0064] In summary, the embodiment of the present application first obtains the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the two. Finally, a classifier is used to obtain a classification result to determine whether to issue a fire safety warning for the communication room, thereby reducing the workload of manual monitoring, thereby improving the intelligence level and operational efficiency of the system, and improving the overall safety of the communication room.

[0065] As described above, the management system 100 based on the intelligent management cabinet of the communication room according to the embodiment of the present application can be implemented in various terminal devices. In one example, the management system 100 based on the intelligent management cabinet of the communication room can be integrated into the terminal device as a software module and / or a hardware module. For example, the management system 100 based on the intelligent management cabinet of the communication room can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the management system 100 based on the intelligent management cabinet of the communication room can also be one of the many hardware modules of the terminal device.

[0066] Alternatively, in another example, the management system 100 of the intelligent management cabinet based on the communication room and the terminal device may also be separate devices, and the management system 100 of the intelligent management cabinet based on the communication room may be connected to the terminal device through a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0067] Figure 5 FIG. 1 is a flow chart of a management method of an intelligent management cabinet based on a communication room according to an embodiment of the present application. Figure 5 As shown, according to the management method of the intelligent management cabinet based on the communication room according to the embodiment of the present application, it includes: S110, obtaining the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor; S120, extracting the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet from the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor; S130, judging whether to issue a fire safety warning for the communication room based on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet.

[0068] Here, those skilled in the art can understand that the specific operations of each step in the management method of the intelligent management cabinet based on the communication room have been referred to above. Figures 1 to 4The management system of the intelligent management cabinet based on the communication room has been introduced in detail, and therefore, its repeated description will be omitted.

[0069] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application.

[0070] See also Figure 6 , Figure 6 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0071] The processor 11 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in the embodiment of the present application; the memory 12 can be implemented in the form of a read-only memory (Read Only Memory, ROM), a static storage device, a dynamic storage device or a random access memory (Random Access Memory, RAM), etc.

[0072] The memory 12 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 12, and the processor 11 calls and executes the management method of the intelligent management cabinet based on the communication room of the embodiment of the present application;

[0073] Input / output interface 13, used to implement information input and output;

[0074] The communication interface 14 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0075] A bus 15 that transmits information between the various components of the device (e.g., the processor 11, the memory 12, the input / output interface 13, and the communication interface 14);

[0076] The processor 11 , the memory 12 , the input / output interface 13 and the communication interface 14 are connected to each other in communication within the device via a bus 15 .

[0077] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the management method of the intelligent management cabinet based on the communication room is implemented.

[0078] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0079] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0080] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0081] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0082] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0083] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0084] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0085] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0086] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit 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 this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs. The preferred embodiment of the embodiment of the present application is described above with reference to the accompanying drawings, and the scope of rights of the embodiment of the present application is not limited thereto. Any modifications, equivalent substitutions and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of rights of the embodiments of the present application.

Claims

1. A management system based on an intelligent management cabinet in a communication room, characterized in that: include: The communication room management cabinet data acquisition module is used to obtain the internal monitoring video of the communication room management cabinet collected by the camera and the smoke value of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor; A communication room management cabinet data extraction module is used to extract the communication room management cabinet internal monitoring focus feature vector and the communication room management cabinet smoke time series correlation feature vector from the communication room management cabinet internal monitoring video collected by the camera and the communication room management cabinet environment smoke values ​​at multiple predetermined time points collected by the smoke sensor; The fire safety warning judgment module is used to determine whether to issue a fire safety warning for the communication room based on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet.

2. The management system of the intelligent management cabinet based on the communication room according to claim 1 is characterized in that: The communication room management cabinet data extraction module includes: A management cabinet internal monitoring feature extraction unit, used for extracting features from the communication room management cabinet internal monitoring video collected by the camera to obtain the communication room management cabinet internal monitoring focus feature vector; The management cabinet environment smoke feature extraction unit is used to extract features from the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor to obtain the communication room management cabinet smoke time series correlation feature vector.

3. The management system of the intelligent management cabinet based on the communication room according to claim 2 is characterized in that: The management cabinet internal monitoring feature extraction unit includes: The management cabinet internal monitoring data preprocessing subunit is used to perform data preprocessing on the communication room management cabinet internal monitoring video collected by the camera to obtain multiple communication room management cabinet internal monitoring enhanced key frames; The management cabinet internal monitoring feature encoding subunit is used to perform feature encoding on the multiple communication room management cabinet internal monitoring enhanced key frames to obtain the communication room management cabinet internal monitoring focus feature vector.

4. The management system of the intelligent management cabinet based on the communication room according to claim 3 is characterized in that: The management cabinet internal monitoring data preprocessing subunit includes: Extracting a plurality of key frames of internal monitoring of the management cabinet of the communication room from the internal monitoring video of the management cabinet of the communication room collected by the camera; The multiple communication room management cabinet internal monitoring key frames are passed through the communication room management cabinet internal monitoring convolutional neural network as a noise reducer to obtain the multiple communication room management cabinet internal monitoring enhanced key frames.

5. The management system of the intelligent management cabinet based on the communication room according to claim 4 is characterized in that: The management cabinet internal monitoring feature coding subunit includes: The plurality of communication room management cabinet internal monitoring enhanced key frames are passed through a communication room management cabinet internal monitoring interest region feature encoder based on a target detection network to obtain a plurality of communication room management cabinet internal monitoring interest regions; Arrange the plurality of communication room management cabinet internal monitoring interest regions along the sample dimension to obtain a communication room management cabinet internal monitoring interest input tensor; The communication room management cabinet internal monitoring interest input tensor is passed through the communication room management cabinet internal monitoring interest feature encoder based on a three-dimensional convolutional neural network to obtain the communication room management cabinet internal monitoring interest feature vector.

6. The management system of the intelligent management cabinet based on the communication room according to claim 5 is characterized in that: The management cabinet environment smoke feature extraction unit includes: Arranging the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor into multiple management cabinet smoke input vectors; The multiple management cabinet smoke input vectors are passed through a communication room management cabinet smoke timing encoder to obtain the communication room management cabinet smoke timing associated feature vector.

7. The management system of the intelligent management cabinet based on the communication room according to claim 6 is characterized in that: The fire safety early warning judgment module includes: A communication room management cabinet safety feature fusion unit is used to perform perspective conversion matching based on topological correlation correction on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet to obtain a fire safety early warning judgment classification feature vector; The communication room fire safety warning classification judgment unit is used to pass the fire safety warning judgment classification feature vector through a classifier to obtain a classification result, and the classification result is used to determine whether to issue a communication room fire safety warning.

8. The management system of the intelligent management cabinet based on the communication room according to claim 7 is characterized in that: The communication room management cabinet security feature fusion unit is used to: Constructing a communication room management cabinet internal monitoring-smoke time series distance topological matrix between the communication room management cabinet internal monitoring attention feature vector and the communication room management cabinet smoke time series associated feature vector; Constructing a communication room management cabinet internal monitoring-smoke time series feature granularity interaction matrix between the communication room management cabinet internal monitoring focus feature vector and the communication room management cabinet smoke time series associated feature vector; Based on the internal monitoring-smoke time series distance topological matrix of the communication room management cabinet, the internal monitoring-smoke time series feature granularity interaction matrix of the communication room management cabinet is topologically corrected to obtain the internal monitoring-smoke time series topological correction association matrix of the communication room management cabinet; Taking the internal monitoring-smoke timing topology correction association matrix of the communication room management cabinet as the perspective co-projection space, projecting the internal monitoring attention feature vector of the communication room management cabinet and the smoke timing association feature vector of the communication room management cabinet to the perspective co-projection space to obtain the perspective conversion communication room management cabinet internal monitoring attention feature vector and the perspective conversion communication room management cabinet smoke timing association feature vector; The fire safety early warning judgment classification feature vector is obtained by fusing the internal monitoring focus feature vector of the communication room management cabinet of the perspective conversion and the smoke time series association feature vector of the communication room management cabinet of the perspective conversion.

9. A management method for an intelligent management cabinet based on a communication room, characterized in that: include: Obtaining the internal monitoring video of the management cabinet of the communication room collected by the camera and the smoke values ​​of the management cabinet environment of the communication room collected by the smoke sensor at multiple predetermined time points; Extracting the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet from the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor; Based on the internal monitoring focus feature vector of the communication room management cabinet and the smoke time series association feature vector of the communication room management cabinet, it is determined whether to issue a fire safety warning for the communication room.

10. The management method of the intelligent management cabinet based on the communication room according to claim 9 is characterized in that: Extracting the internal monitoring feature vector of the communication room management cabinet and the smoke time series correlation feature vector of the communication room management cabinet from the internal monitoring video of the communication room management cabinet collected by the camera and the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor, including: Performing feature extraction on the internal monitoring video of the communication room management cabinet collected by the camera to obtain a monitoring focus feature vector of the internal monitoring of the communication room management cabinet; Feature extraction is performed on the smoke values ​​of the communication room management cabinet environment at multiple predetermined time points collected by the smoke sensor to obtain the communication room management cabinet smoke time series correlation feature vector.

Images