An intelligent management method and system based on spatial merging

Abstract

The present invention provides an intelligent management method and system based on spatial merging, which relates to the technical field of data processing. The method includes calibrating the types of monitoring data for different monitoring spaces to form parameter-based monitoring spaces and image-based monitoring spaces; obtaining the historical monitoring data of the monitoring spaces, and performing feature division based on the safety state on the monitoring spaces according to the historical monitoring data to form safety state reference monitoring data; obtaining the historical monitoring data of the monitoring spaces, and performing feature division based on the early warning state to form early warning state comparison monitoring data; obtaining the real-time monitoring data of the monitoring spaces, performing early warning monitoring analysis based on the state to form early warning monitoring analysis result data; and performing monitoring early warning according to the early warning monitoring analysis result data. This method realizes efficient and accurate real-time early warning monitoring through early warning monitoring analysis using Internet of Things data, and further optimizes spatial monitoring management.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent management method and system based on space merging. Background Art

[0002] The development of IoT technology has promoted technological innovation in all walks of life. In the field of property management, various types of sensor systems and monitoring systems are set up because of the need to reasonably monitor different spaces. However, these systems are different in type and brand, which makes it impossible to form a unified management, thereby increasing the complexity and difficulty of property management. IoT technology may become a technical means to overcome this difficulty.

[0003] At present, considering improving the service quality of property and reducing the cost of property services, adopting a unified management platform to achieve regional property service management can not only improve the quality of property services, but also bring a more efficient and cost-saving work management model to enterprises. At present, one of the main purposes of using the Internet of Things system for unified automated and efficient space monitoring is to warn the real-time status of the space, and then provide early warning information to all parties in a timely manner to ensure timely and efficient processing of the early warning status.

[0004] Therefore, it is an urgent problem to design an intelligent management method and system based on space merging, to achieve efficient and accurate real-time early warning monitoring by utilizing IoT data for early warning monitoring and analysis, and then to optimize space monitoring management. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent management method based on space merging, which distinguishes different types of space monitoring by type calibration of space monitoring data, and then adopts different early warning monitoring and analysis methods to realize efficient and accurate real-time early warning monitoring of space. Among them, for different types of monitoring data, comparative data for judging the safety state and the early warning state are formed on the basis of using historical monitoring data, so that the early warning state can be quickly monitored and judged when obtaining real-time monitoring data. For the judgment of the safety state, the reference monitoring data of the safety state takes into account the sudden characteristics of the early warning state. In most cases, the space state is in the safety state. Therefore, in order to avoid the analysis and processing of the real-time data in the early warning state, which leads to excessive data analysis processing and reduces the efficiency of early warning monitoring analysis, a simple load is provided for the safety state, which further improves the efficiency and accuracy of early warning analysis. Reasonable space monitoring optimization is realized from the early warning analysis mode and property management mode, and different spaces in the monitoring area can be unified for real-time early warning monitoring, providing accurate and timely early warning information for all parties related to the property, and improving the quality of property management.

[0006] The purpose of the present invention is also to provide an intelligent management system based on space merging, which realizes the collection of monitoring data of different types and different spaces through different types of monitoring data collection units, and provides an important data basis for the subsequent early warning monitoring analysis. At the same time, the analysis unit can uniformly use the historical data and real-time data collected by the collection unit to perform real-time monitoring and analysis of the early warning status, which is efficient and accurate, and use the early warning information processing unit to provide the necessary feature information of the early warning information to all parties related to the property in a timely manner. The entire system realizes efficient and accurate real-time analysis of early warning monitoring under space integration through simple and tight functional unit configuration, which improves the quality of the system for property management and the quality of property services.

[0007] In the first aspect, the present invention provides an intelligent management method based on space merging, including calibrating the monitoring data types of different monitoring spaces to form parameter-type monitoring spaces and image-type monitoring spaces; for different types of monitoring spaces, obtaining historical monitoring data of the monitoring spaces, and performing feature division of the monitoring spaces based on the safety status according to the historical monitoring data to form safety status reference monitoring data; for different types of monitoring spaces, obtaining historical monitoring data of the monitoring spaces, and performing feature division based on the warning status to form warning status comparison monitoring data; obtaining real-time monitoring data of the monitoring spaces, and performing state-based warning monitoring analysis in combination with the safety status reference monitoring data and the warning status comparison monitoring data to form warning monitoring analysis result data; and performing monitoring and warning according to the warning monitoring analysis result data.

[0008] In the present invention, the method distinguishes different types of space monitoring by type calibration of space monitoring data, and then adopts different early warning monitoring and analysis methods to realize efficient and accurate real-time early warning monitoring of space. Among them, for different types of monitoring data, comparative data for judging the safety state and the early warning state are formed on the basis of using historical monitoring data, so that the monitoring and judgment of the early warning state can be quickly performed when obtaining real-time monitoring data. For the judgment of the safety state, the reference monitoring data of the safety state takes into account the sudden characteristics of the early warning state. In most cases, the space state is in the safety state. Therefore, in order to avoid the analysis and processing of the real-time data being consistently in the early warning state, which leads to excessive data analysis processing and reduces the efficiency of early warning monitoring and analysis, a simple load is provided for the safety state, which further improves the efficiency and accuracy of the early warning analysis. Reasonable space monitoring optimization is achieved from the early warning analysis mode and the property management mode, and different spaces in the monitoring area can be unified for real-time early warning monitoring, providing accurate and timely early warning information for all parties related to the property, and improving the quality of property management.

[0009] As a possible implementation, for different types of monitored spaces, historical monitoring data of the monitored space is obtained, and based on the historical monitoring data, feature partitioning of the monitored space is performed based on the safety state to form safety state reference monitoring data, including: for parameter-type monitored spaces, historical parameter monitoring data is obtained; a parameter monitoring period is set, and based on the parameter monitoring period as the basis for data partitioning, the historical parameter monitoring data in the historical parameter monitoring data that is not marked as the warning state is extracted segment by segment in the order of the time dimension to form a periodic non-warning parameter monitoring information set A; for each parameter in the periodic non-warning parameter monitoring information set A, the numerical range on all parameter monitoring periods is obtained, and the intersection operation is performed on the numerical ranges to determine the safety state range corresponding to each parameter; the safety state ranges corresponding to all parameters in the periodic non-warning parameter monitoring information set A are aggregated to form a parameter range safety state monitoring set B.

[0010] In the present invention, here feature partitioning based on the safety state is performed for the parameter-type monitored space. It should be noted that the parameter-type monitored space is a space that monitors target parameters in the monitored space such as equipment operation status, temperature and humidity monitoring, etc. Usually, the monitoring of these parameters is carried out continuously. Therefore, in order to perform reasonable early warning analysis, it is necessary to perform reasonable segmented partitioning of the parameter data acquisition, and then realize comparative analysis to extract the parameter numerical feature information in the safety state. Considering that the monitored parameters in different parameter-type monitored spaces are different, the parameter monitoring period can be set according to the time characteristics of the monitored parameters, so as to realize reasonable partitioning of the parameter numerical data. In addition, for obtaining the range of parameters in the safety state, considering that under normal circumstances, the parameters are within a certain reasonable range, and all the target parameters to be monitored in the space should be in a normal state. Occasionally, there will be abnormal fluctuations in some parameters, but this kind of abnormality does not necessarily form a warning state. Therefore, in order to ensure that the real-time state of the space can be accurately determined, the union of the same-type numerical ranges on the parameter monitoring period is performed for all the parameters to be monitored in the space to form a relatively narrow range. It can be understood that the abnormal fluctuations of the parameters are actually sporadic and low-frequency in the entire time dimension of the space monitoring. Therefore, the relatively narrow range generated by the union can also realize the state judgment of most monitoring moments, which also performs a large amount of data screening for the early warning analysis, reduces the data input volume for the subsequent early warning analysis, and improves the efficiency of the early warning analysis.

[0011] As a possible implementation, for different types of monitored spaces, historical monitoring data of the monitored space is obtained, and feature division based on the warning state is performed to form warning state comparison monitoring data, including: based on the parameter monitoring period as the basis for data division, segment extraction in the order of the time dimension is performed on the parameter type historical monitoring data marked with the warning state in the parameter type historical monitoring data to form a periodic warning parameter monitoring information set C; for each parameter in the periodic warning parameter monitoring information set C, the numerical range on all parameter monitoring periods is obtained, and the union operation is performed on the numerical ranges to determine the warning state range corresponding to each parameter; the warning state ranges corresponding to all parameters in the periodic warning parameter monitoring information set C are aggregated to form a parameter range warning state monitoring set D.

[0012] In the present invention, it can be understood that when the monitored space is in a warning state, the values of the monitored parameters will fluctuate greatly and involve multiple parameters. Therefore, first, the union calculation of the numerical ranges of the parameters in the warning state is performed, and a relatively broad numerical range of the parameters that can be used for warning judgment can be obtained. In this way, the complete coverage of the change in the numerical range of the parameters marked in the warning state can be fully realized, and the accuracy of warning state analysis can be improved.

[0013] As a possible implementation, real-time monitoring data of the monitored space is obtained, and state-based warning monitoring analysis is performed in combination with safety state reference monitoring data and warning state comparison monitoring data to form warning monitoring analysis result data, including: taking the parameter monitoring period as the time range for real-time data acquisition, periodically obtaining real-time parameter monitoring data, and determining the real-time parameter numerical range corresponding to each parameter in the real-time parameter monitoring data; the following comparison analysis is performed on the real-time parameter numerical range of each parameter in the real-time parameter monitoring data and the safety state range of the corresponding parameter in the parameter range safety state monitoring set B: if the real-time parameter numerical ranges of all parameters belong to the corresponding safety state ranges, it is determined that the parameter type monitored space is in a safety monitoring state; otherwise, it is determined that it is in a non-safety monitoring state. For the parameter type monitored space in a non-safety monitoring state, a warning parameter quantity threshold α is set, and the following comparison judgment is performed in combination with the warning state range corresponding to each parameter in the parameter range warning state monitoring set D: if the total number of parameters whose real-time parameter numerical ranges belong to the corresponding warning state ranges is less than the warning parameter quantity threshold α, it is judged that the parameter type monitored space is in a non-warning state; if the total number of parameters whose real-time parameter numerical ranges belong to the corresponding warning state ranges is greater than or equal to the warning parameter quantity threshold α, it is judged that the parameter type monitored space is in a warning state.

[0014] In the present invention, after establishing the numerical range in the parameter safe state and the data range in the warning state, the state of the current monitoring space can be judged by combining the real-time numerical range. For the monitoring space that is not in the safe monitoring state, it is possible that the abnormality in its parameter part is only an occasional fluctuation. Therefore, it is necessary to count the number of parameters as the basis for judging whether the monitoring space is in the warning state. It should be noted that for the warning parameter quantity threshold, considering the correlation between different parameters in the monitoring space, it is impossible to independently judge whether the monitoring space is in the warning state based on whether a single parameter exceeds the range. Therefore, the warning parameter quantity threshold is an index that takes into account the correlation between different parameters. For example, for the operation of the monitoring device, a change in a single current exceeding the range cannot clearly judge that the device is in the warning state, but the changes in parameters such as voltage and power related to the current exceeding the range together can clearly indicate that the device is in the warning state. Therefore, the size of the warning parameter quantity threshold can be set according to the tightness of the connection between different parameters in the monitoring space, or based on some golden parameter indicators.

[0015] As a possible implementation, according to the warning monitoring analysis result data, monitoring and warning are carried out, including: determining the parameter type monitoring space in the warning state, calibrating the parameters whose real-time parameter numerical range belongs to the corresponding warning state range to form warning abnormal parameters; combining the warning abnormal parameters and the real-time parameter numerical range corresponding to the warning abnormal parameters to form monitoring and warning information.

[0016] In the present invention, after determining the warning state, in order to enable all parties with different requirements to obtain specific and accurate warning information, it is essential to include the real-time range of parameters and the warning object as part of the warning information.

[0017] As a possible implementation, for different types of monitored spaces, historical monitoring data of the monitored space is obtained, and based on the historical monitoring data, feature partitioning of the monitored space based on the safety state is performed to form safety state reference monitoring data, including: for image-based monitored spaces, historical image monitoring data is obtained; an image monitoring period is set, and based on the image monitoring period as the data partitioning basis, the historical image monitoring data that has not been marked as the warning state in the historical image monitoring data is extracted in segments according to the time dimension order to form a periodic non-warning image monitoring information set E; a safety and stability duration threshold β is set. For all images in the periodic non-warning image monitoring information set E, the following duration analysis and judgment based on the object are performed on the images: if the duration of the same object existing in each image monitoring period in all images is greater than or equal to the safety and stability duration threshold β, the object is determined to be a safe and stable object; if the duration of the object existing in the corresponding image monitoring period in all images is less than the safety and stability duration threshold β, the object is determined to be an unsafe and stable object; all safe and stable objects are aggregated to form an image object safety state monitoring set F.

[0018] In the present invention, for image-based monitored spaces, different from parameter-based monitored spaces, the image data obtained by the image-based monitored spaces can integrally contain all state-related contents of the space. Therefore, when performing data extraction for the safety state, it is necessary to be based on the images. Of course, the acquisition of image data is also continuous. Therefore, it is necessary to set the image monitoring period as the data partitioning basis according to the needs of analysis or the time characteristics of the acquisition of image data. For image data, the image data in the warning state must be that the objects in the image have changed. By extracting and analyzing the objects in the image, the state of the monitored space can be accurately grasped. It should be noted that the extraction of objects in the image can be based on the extraction of the object boundary range, or the extraction of data on the position and object area, or the identification and extraction based on color, etc., which can be determined according to actual needs. It can be understood that for a monitored space in a safe state, in fact, the objects in the image are stable and unchanged for a long time. For example, for the warning situation of monitoring a water pool without people falling into it and drowning, the environment around the water pool is stable and unchanged for a long time, while the people near the water pool are changes that exist briefly. Therefore, using the duration of the object's existence as the judgment of whether it is a safe and stable object is a simple and accurate judgment method.

[0019] As a possible implementation, for different types of monitoring spaces, historical monitoring data of the monitoring space is obtained, and feature division based on the warning state is performed to form warning state comparison monitoring data, including: based on the image monitoring cycle as the basis for data division, the image-type historical monitoring data marked with the warning state in the image-type historical monitoring data is extracted segment by segment in the order of the time dimension to form a periodic warning image monitoring information set G; a warning duration threshold η and a warning quantity threshold γ are set, and for all images in the periodic warning image monitoring information set G, after removing the safe and stable objects in the image object safety state monitoring set F, the following analysis and judgment are performed: if there is an object whose duration in the image monitoring cycle is greater than or equal to the warning duration threshold η, and the number of times the object appears in different image monitoring cycles is greater than or equal to the warning quantity threshold γ, then the object is determined as a warning object; otherwise, it is determined as a non-warning object; all warning objects are aggregated to form an image object warning state monitoring set H.

[0020] In the present invention, for obtaining the warning objects of images, it is necessary to obtain the object information existing in most of the warning images in the warning image information. This extraction method of warning objects can also determine the positions and events in the monitoring space where warning situations are likely to occur. Of course, for warning objects, a certain duration is also required to accurately determine. After all, some are occasional events or warning states that can weaken by themselves, and the continuously intensifying warnings are the most worthy of attention.

[0021] As a possible implementation, real-time monitoring data of the monitoring space is obtained, and state-based early warning monitoring analysis is carried out by combining safety state reference monitoring data and early warning state comparison monitoring data to form early warning monitoring analysis result data, including: taking the image monitoring period as the time range for real-time data acquisition, periodically obtaining real-time image monitoring data, and determining real-time image objects existing in each image in the real-time image monitoring data; comparing all real-time image objects in the real-time image monitoring data with the safe and stable objects in the image object safety state monitoring set F as follows: if all real-time image objects belong to the safe and stable objects in the image object safety state monitoring set F, it is determined that the image class monitoring space is in a safe monitoring state; otherwise, it is determined that it is in an unsafe monitoring state. For the image class monitoring space in an unsafe monitoring state, an early warning duration threshold T is set, and the following comparison and judgment are carried out in combination with all warning objects in the image object early warning state monitoring set H: if there is a real-time image object belonging to the warning objects in the image object early warning state monitoring set H, but the duration of its existence in the corresponding image monitoring period is less than the early warning duration threshold T, it is judged that the image class monitoring space is in a non-warning state; if there is a real-time image object belonging to the warning objects in the image object early warning state monitoring set H, but the duration of its existence in the corresponding image monitoring period is greater than or equal to the early warning duration threshold T, it is judged that the image class monitoring space is in a warning state.

[0022] In the present invention, by first using safety state data to compare the real-time state of the monitoring space, the processing amount of continuously comparing real-time data with early warning state data directly can be reduced. After all, in most cases, the monitoring space is in a safe state. The analysis of early warning monitoring after screening the early warning data can also improve the accuracy of the analysis.

[0023] As a possible implementation, according to the early warning monitoring analysis result data, monitoring early warning is carried out, including: determining the image class monitoring space in a warning state, calibrating the real-time image objects belonging to the warning objects in the image object early warning state monitoring set H and having a duration of existence greater than or equal to the early warning duration threshold T in the corresponding image monitoring period to form warning abnormal objects; combining the warning abnormal objects and the duration of the warning abnormal objects to form monitoring early warning information.

[0024] In the present invention, after determining the warning state, in order to enable all parties with different requirements to obtain specific and accurate early warning information, it is essential to take the warning objects and the duration of the warning objects as part of the early warning information.

[0025] Second aspect, the present invention provides an intelligent management system based on space merging, which is applied to the intelligent management method based on space merging described in the first aspect, and includes a monitoring space numerical data acquisition unit for collecting the parameter numerical range in the parameter type monitoring space to form parameter type historical monitoring data and real-time parameter monitoring data; a monitoring space image data acquisition unit for collecting images in the image type monitoring space to form image type historical monitoring data and real-time image monitoring data; an early warning analysis unit for obtaining the parameter type historical monitoring data of the monitoring space numerical data acquisition unit and the image type historical monitoring data of the monitoring space image data acquisition unit to form corresponding safety state reference monitoring data and early warning state comparison monitoring data, and for obtaining the real-time parameter monitoring data of the monitoring space numerical data acquisition unit and the real-time image monitoring data of the monitoring space image data acquisition unit to perform state-based early warning monitoring analysis to form early warning monitoring analysis result data; an early warning information processing unit for obtaining the early warning monitoring analysis result data formed by the early warning analysis unit to form early warning information and sending it to different monitoring objects.

[0026] In the present invention, the system realizes the acquisition of monitoring data of different types and different spaces through the acquisition units of different types of monitoring data, providing an important data basis for the subsequent analysis of early warning monitoring. At the same time, the analysis unit can uniformly use the historical data and real-time data collected by the acquisition unit to perform real-time monitoring analysis of the early warning state, which is efficient and accurate, and uses the early warning information processing unit to timely provide the necessary characteristic information of the early warning information to all parties related to the property management. The entire system realizes the real-time analysis of early warning monitoring under efficient and accurate space integration through simple and tight functional unit configuration, improving the quality of the system for property management and enhancing the quality of property services.

[0027] The beneficial effects of the intelligent management method and system based on space merging provided by the present invention are:

[0028] This method differentiates different types of space monitoring by calibrating the types of space monitoring data, and then adopts different early warning monitoring analysis methods accordingly to achieve efficient and accurate real-time space early warning monitoring. Among them, for different types of monitoring data, comparison data for judging the safety state and early warning state are formed based on historical monitoring data, so as to quickly judge the early warning state when real-time monitoring data is obtained. The reference monitoring data for the safety state is used as the judgment of the safety state considering the sudden characteristics of the early warning state. In most cases, the space state is in a safe state. Therefore, to avoid excessive data analysis and processing due to the real-time data always being in the early warning state analysis and processing, a simple load for the safety state is provided, which further improves the efficiency and accuracy of the early warning analysis. From the early warning analysis mode and property management method, reasonable space monitoring optimization is achieved, and different spaces in the monitoring area can be unified for real-time early warning monitoring, providing accurate and timely early warning information for all parties related to the property and improving the quality of property management.

[0029] This system realizes the collection of monitoring data for different types and different spaces through the collection units of different types of monitoring data, providing an important data basis for the subsequent analysis of early warning monitoring. At the same time, the analysis unit can uniformly use the historical data and real-time data collected by the collection unit for real-time monitoring and analysis of the early warning state, which is efficient and accurate, and uses the early warning information processing unit to timely provide the necessary characteristic information of the early warning information to all parties related to the property. The entire system realizes the real-time analysis of early warning monitoring under efficient and accurate space integration through simple and compact functional unit configuration, improves the quality of the system for property management, and enhances the quality of property services. Brief Description of the Drawings

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments of the present invention. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0031] Figure 1 It is a step diagram of the intelligent management method based on space merging provided by the embodiment of the present invention. Detailed Embodiments

[0032] The following will describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention.

[0033] The development of Internet of Things technology has promoted technological innovation in various industries. In the property management field, due to the need to reasonably monitor different spaces, various types of sensing systems and monitoring systems are set up. However, due to different types and brands of these systems, it is impossible to form a unified management, which further increases the complexity and difficulty of property management. Internet of Things technology may become a technical means to break through this difficulty.

[0034] Currently, considering improving the service quality of property management and reducing the service cost of property management, adopting a unified management platform to realize regional property service management can not only improve the quality of property service, but also bring a more efficient and cost-saving work management mode for enterprises. At present, for the same automated and efficient space monitoring using the Internet of Things system, one of the main purposes is to give early warnings about the real-time state of the space, and then timely provide early warning information to inform all parties to ensure timely and efficient handling of the early warning state.

[0035] Reference Figure 1 , an embodiment of the present invention provides an intelligent management method based on space merging. This method distinguishes different space monitoring types by calibrating the types of space monitoring data, and then specifically adopts different early warning monitoring analysis methods to achieve efficient and accurate real-time space early warning monitoring. Among them, for different types of monitoring data, comparison data for judging the safety state and early warning state are formed based on historical monitoring data, so as to quickly monitor and judge the early warning state when real-time monitoring data is obtained. For the safety state reference monitoring data used as the judgment of the safety state, considering the sudden characteristics of the early warning state, in most cases, the space state is in a safe state. Therefore, in order to avoid excessive data analysis and processing due to the real-time data always being in the early warning state analysis and processing, a simple load of the safety state is provided, which further improves the efficiency and accuracy of early warning analysis. From the early warning analysis mode and property management method, reasonable space monitoring optimization is realized, and different spaces in the monitoring area can be unified for real-time early warning monitoring, providing accurate and timely early warning information for all parties related to property management and improving the quality of property management.

[0036] The intelligent management method based on space merging specifically includes the following steps:

[0037] S1: Calibrate the types of monitoring data for different monitoring spaces to form parameter-based monitoring spaces and image-based monitoring spaces.

[0038] The type division of the monitoring space is based on the monitoring data, so that reasonable early warning analysis can be carried out for different data types to complete the early warning monitoring in space merging management.

[0039] S2: For different types of monitored spaces, obtain the historical monitoring data of the monitored space, and based on the historical monitoring data, perform a feature division of the monitored space based on the safety state to form safety state reference monitoring data.

[0040] Among them, for different types of monitored spaces, obtaining the historical monitoring data of the monitored space, and based on the historical monitoring data, performing a feature division of the monitored space based on the safety state to form safety state reference monitoring data includes: for parameter-type monitored spaces, obtaining historical parameter monitoring data; setting a parameter monitoring period, using the parameter monitoring period as the basis for data division, and segmenting and extracting the historical parameter monitoring data that is not marked as the warning state in the historical parameter monitoring data in the order of the time dimension to form a periodic non-warning parameter monitoring information set A; for each parameter in the periodic non-warning parameter monitoring information set A, obtaining the numerical range on all parameter monitoring periods, and performing an intersection operation on the numerical ranges to determine the safety state range corresponding to each parameter; aggregating the safety state ranges corresponding to all parameters in the periodic non-warning parameter monitoring information set A to form a parameter range safety state monitoring set B.

[0041] Here, a feature division based on the safety state is performed for the parameter-type monitored space. It should be noted that the parameter-type monitored space is a space that monitors target parameters in the monitored space, such as equipment operating status, temperature and humidity monitoring, etc. Usually, the monitoring of these parameters is carried out continuously. Therefore, in order to perform reasonable early warning analysis, it is necessary to reasonably segment the data collection of the parameters, and then realize the comparative analysis to extract the parameter numerical feature information in the safety state. Considering that the monitored parameters in different parameter-type monitored spaces are different, the parameter monitoring period can be set according to the time characteristics of the monitored parameters, so as to realize reasonable parameter numerical data division. In addition, for obtaining the range of parameters in the safety state, considering that under normal circumstances, the parameters are within a certain reasonable range, and all the target parameters to be monitored in the space should be in a normal state. Occasionally, there will be abnormal fluctuations in some parameters, but this kind of abnormality does not necessarily form a warning state. Therefore, in order to ensure the accurate determination of the real-time state of the space, the union of the same type of numerical ranges on the parameter monitoring period is performed for all the parameters to be monitored in the space to form a relatively narrow range. It can be understood that the abnormal fluctuations of the parameters are actually sporadic and low-frequency in the entire time dimension of the space monitoring. Therefore, the relatively narrow range generated by the union can also realize the state judgment of most monitoring moments, which also screens a large amount of data for the early warning analysis, reduces the data input volume for the subsequent early warning analysis, and improves the efficiency of the early warning analysis.

[0042] For the image-based monitoring space, for different types of monitoring spaces, obtain the historical monitoring data of the monitoring space, and perform feature division based on the safety status on the monitoring space according to the historical monitoring data to form safety status reference monitoring data, including: for the image-based monitoring space, obtain the historical image monitoring data; set the image monitoring period, and use the image monitoring period as the basis for data division to segment and extract the historical image monitoring data that is not marked as the warning state in the historical image monitoring data in the order of the time dimension to form a periodic non-warning image monitoring information set E; set the safety and stability duration threshold β, and for all images in the periodic non-warning image monitoring information set E, perform the following duration analysis and judgment on the images based on the object: if the duration of the same object existing in each image monitoring period in all images is greater than or equal to the safety and stability duration threshold β, determine that the object is a safe and stable object; if the duration of the object existing in the corresponding image monitoring period in all images is less than the safety and stability duration threshold β, determine that the object is an unsafe and stable object; collect all safe and stable objects to form an image object safety status monitoring set F.

[0043] For the image-based monitoring space, different from the parameter-based monitoring space, the image data obtained by the image-based monitoring space can integrally contain all the content related to the state of the space. Therefore, when performing data extraction for the safety status, it needs to be based on the image. Of course, the acquisition of image data is also continuous, so it is necessary to set the image monitoring period as the basis for data division according to the needs of analysis or the time characteristics of the acquisition of image data. For image data, the image data in the warning state must be that the object in the image has changed. By extracting and analyzing the object in the image, the state of the monitoring space can be accurately grasped. It should be noted that the extraction of the object in the image can be based on the extraction of the object boundary range, or the extraction of data on the position and object area, or the identification and extraction based on color, etc., which can be determined according to actual needs. It can be understood that for the monitoring space in a safe state, in fact, the object in the image is stable and unchanged for a long time. For example, for the warning situation of monitoring a water pool without people falling into it and drowning, the environment around the water pool is stable and unchanged for a long time, while the people near the water pool are transient changes. Therefore, using the duration of the object's existence as the judgment of whether it is a safe and stable object is a simple and accurate judgment method.

[0044] S3: For different types of monitoring spaces, obtain the historical monitoring data of the monitoring space and perform feature division based on the warning state to form warning state comparison monitoring data.

[0045] Among them, for the parameter - type monitoring space: based on the parameter monitoring period as the basis for data division, segment - extract the parameter - type historical monitoring data marked with the warning state in the parameter - type historical monitoring data in the order of the time dimension to form the periodic warning parameter monitoring information set C; for each parameter in the periodic warning parameter monitoring information set C, obtain the value ranges on all parameter monitoring periods, and perform a union operation on the value ranges to determine the warning - state range corresponding to each parameter; aggregate the warning - state ranges corresponding to all parameters in the periodic warning parameter monitoring information set C to form the parameter - range warning - state monitoring set D.

[0046] It can be understood that when the monitoring space is in a warning state, the values of the monitored parameters will fluctuate greatly and involve multiple parameters. Therefore, first perform a union calculation on the value ranges of the parameters in the warning state to obtain a relatively broad parameter value range that can be used as a warning judgment. In this way, it can fully cover the changes in the value ranges of the parameters marked in the warning state and improve the accuracy of warning - state analysis.

[0047] For the image - type monitoring space, for different types of monitoring spaces, obtain the historical monitoring data of the monitoring space and perform feature division based on the warning state to form the warning - state comparison monitoring data, including: based on the image monitoring period as the basis for data division, segment - extract the image - type historical monitoring data marked with the warning state in the image - type historical monitoring data in the order of the time dimension to form the periodic warning image monitoring information set G; set the warning - duration threshold η and the warning - quantity threshold γ, and for all images in the periodic warning image monitoring information set G, after removing the safe and stable objects in the image - object safety - state monitoring set F, perform the following analysis and judgment: if there exists an object whose duration in the image monitoring period is greater than or equal to the warning - duration threshold η, and the number of times the object appears in different image monitoring periods is greater than or equal to the warning - quantity threshold γ, then determine the object as a warning object; otherwise, determine it as a non - warning object; aggregate all warning objects to form the image - object warning - state monitoring set H.

[0048] For obtaining the warning objects of images, it is necessary to obtain the object information that exists in most warning images in the warning - image information. This way of extracting warning objects can also determine the locations and events in the monitoring space where warning situations are likely to occur. Of course, for warning objects, a certain duration is also required to be accurately determined. After all, some are occasional events or warning states that can weaken by themselves, and the continuously intensifying warnings are the most worthy of attention.

[0049] S4: Obtain the real - time monitoring data of the monitoring space, and perform state - based warning - monitoring analysis by combining the safety - state reference monitoring data and the warning - state comparison monitoring data to form the warning - monitoring analysis result data.

[0050] Among them, for the parameter type monitoring space, real-time monitoring data of the monitoring space is obtained, and state-based early warning monitoring analysis is carried out by combining the safety state reference monitoring data and the warning state comparison monitoring data to form early warning monitoring analysis result data, including: taking the parameter monitoring period as the time range for real-time data acquisition, periodically obtaining real-time parameter monitoring data, and determining the real-time parameter value range corresponding to each parameter in the real-time parameter monitoring data; comparing and analyzing the real-time parameter value range of each parameter in the real-time parameter monitoring data with the safety state range of the corresponding parameter in the parameter range safety state monitoring set B as follows: if the real-time parameter value ranges of all parameters belong to the corresponding safety state ranges, it is determined that the parameter type monitoring space is in a safety monitoring state; otherwise, it is determined that it is in a non-safety monitoring state. For the parameter type monitoring space in a non-safety monitoring state, an early warning parameter quantity threshold α is set, and combined with the warning state range corresponding to each parameter in the parameter range warning state monitoring set D, the following comparison and judgment are carried out: if the total number of parameters whose real-time parameter value ranges belong to the corresponding warning state ranges is less than the early warning parameter quantity threshold α, it is judged that the parameter type monitoring space is in a non-warning state; if the total number of parameters whose real-time parameter value ranges belong to the corresponding warning state ranges is greater than or equal to the early warning parameter quantity threshold α, it is judged that the parameter type monitoring space is in a warning state.

[0051] After establishing the numerical range in the parameter safety state and the data range in the warning state, the state of the current monitoring space can be judged by combining the real-time numerical range. For the monitoring space that is not in the safety monitoring state, the abnormality of its parameter part may only be an occasional fluctuation, so it is necessary to count the number of parameters as the basis for judging whether the monitoring space is in the warning state. It should be noted that for the early warning parameter quantity threshold, considering that there will be correlations between different parameters in the monitoring space, it is not possible to independently judge whether the monitoring space is in the warning state based on whether a single parameter exceeds the range. Therefore, the early warning parameter quantity threshold is an index that takes into account the correlations between different parameters. For example, for the operation of monitoring equipment, a single change in current exceeding the range cannot clearly judge that the equipment is in the warning state, but the simultaneous change in parameters such as voltage and power related to the current exceeding the range can clearly indicate that the equipment is in the warning state. Therefore, the size of the early warning parameter quantity threshold can be set according to the tightness of the connection between different parameters in the monitoring space, or based on certain golden parameter indicators.

[0052] For the image-based monitoring space, real-time monitoring data of the monitoring space is obtained, and state-based early warning monitoring analysis is carried out by combining the safety state reference monitoring data and the early warning state comparison monitoring data to form early warning monitoring analysis result data, including: taking the image monitoring period as the time range for real-time data acquisition, periodically obtaining real-time image monitoring data, and determining real-time image objects existing in each image in the real-time image monitoring data; comparing all the real-time image objects in the real-time image monitoring data with the safe and stable objects in the image object safety state monitoring set F as follows: if all the real-time image objects belong to the safe and stable objects in the image object safety state monitoring set F, it is determined that the image-based monitoring space is in a safe monitoring state; otherwise, it is determined that it is in an unsafe monitoring state. For the image-based monitoring space in an unsafe monitoring state, an early warning duration threshold T is set, and combined with all the early warning objects in the image object early warning state monitoring set H, the following comparison and judgment are carried out: if there is a real-time image object belonging to the early warning objects in the image object early warning state monitoring set H, but the duration of its existence in the corresponding image monitoring period is less than the early warning duration threshold T, it is judged that the image-based monitoring space is in a non-early warning state; if there is a real-time image object belonging to the early warning objects in the image object early warning state monitoring set H, but the duration of its existence in the corresponding image monitoring period is greater than or equal to the early warning duration threshold T, it is judged that the image-based monitoring space is in an early warning state.

[0053] For the state that the monitoring space is in real time, first using the safety state data for comparison can reduce the processing volume of continuously comparing the real-time data with the early warning state data directly. After all, in most cases, the monitoring space is in a safe state. The analysis of early warning monitoring after screening the early warning data can also improve the accuracy of the analysis.

[0054] S5: Carry out monitoring and early warning according to the early warning monitoring analysis result data.

[0055] Among them, for the parameter-based monitoring space, according to the early warning monitoring analysis result data, monitoring and early warning are carried out, including: determining the parameter-based monitoring space in the early warning state, calibrating the parameters whose real-time parameter value ranges belong to the corresponding early warning state ranges to form early warning abnormal parameters; combining the early warning abnormal parameters and the real-time parameter value ranges corresponding to the early warning abnormal parameters to form monitoring and early warning information. After determining the early warning state, in order to enable all parties with different requirements to obtain specific and accurate early warning information, it is essential to include the real-time range of the parameters and the early warning objects as part of the early warning information.

[0056] For the image monitoring space, based on the early warning monitoring analysis result data, early warning monitoring is carried out, including: determining the image monitoring space in the early warning state, calibrating the real-time image objects that belong to the early warning object monitoring set H of the image object early warning state and whose existence duration in the corresponding image monitoring period is greater than or equal to the early warning duration threshold T, to form early warning abnormal objects; combining the early warning abnormal objects and the duration of the early warning abnormal objects to form the monitoring early warning information. After determining the early warning state, in order to enable all parties with different requirements to obtain specific and accurate early warning information, it is essential to take the early warning object and the duration of the early warning object as part of the early warning information.

[0057] The present invention also provides an intelligent management system based on space merging. This system adopts the intelligent management method based on space merging provided by the present invention, including a monitoring space numerical data acquisition unit for acquiring the parameter numerical range in the parameter type monitoring space to form parameter type historical monitoring data and real-time parameter monitoring data; a monitoring space image data acquisition unit for acquiring images in the image type monitoring space to form image type historical monitoring data and real-time image monitoring data; an early warning analysis unit for obtaining the parameter type historical monitoring data of the monitoring space numerical data acquisition unit and the image type historical monitoring data of the monitoring space image data acquisition unit to form corresponding safety state reference monitoring data and early warning state comparison monitoring data, and for obtaining the real-time parameter monitoring data of the monitoring space numerical data acquisition unit and the real-time image monitoring data of the monitoring space image data acquisition unit to carry out state-based early warning monitoring analysis to form early warning monitoring analysis result data; an early warning information processing unit for obtaining the early warning monitoring analysis result data formed by the early warning analysis unit to form early warning information and sending it to different monitoring objects.

[0058] This system realizes the acquisition of monitoring data for different types of spaces of different types through the acquisition units of different types of monitoring data, providing an important data basis for the subsequent analysis of early warning monitoring. At the same time, the analysis unit can uniformly use the historical data and real-time data collected by the acquisition units to carry out real-time monitoring analysis of the early warning state, efficiently and accurately, and use the early warning information processing unit to timely provide the necessary characteristic information of the early warning information to all parties related to the property management. The entire system realizes the real-time analysis of early warning monitoring under efficient and accurate space integration through simple and tight functional unit configuration, improving the quality of the system for property management and enhancing the quality of property services.

[0059] In summary, the beneficial effects of the intelligent management method and device based on space merging provided by the embodiments of the present invention are as follows:

[0060] This method differentiates different types of spatial monitoring by calibrating the types of spatial monitoring data, and then adopts different early warning monitoring analysis methods accordingly to achieve efficient and accurate real-time spatial early warning monitoring. Among them, for different types of monitoring data, comparison data for judging the safety state and early warning state are formed based on historical monitoring data, so as to quickly monitor and judge the early warning state when obtaining real-time monitoring data. For the safety state reference monitoring data used as the judgment of the safety state, the sudden characteristics of the early warning state are considered. In most cases, the spatial state is in a safe state. Therefore, to avoid excessive data analysis and processing due to the real-time data always being in the early warning state analysis and processing, a simple load for the safety state is provided, which further improves the efficiency and accuracy of the early warning analysis. From the aspects of the early warning analysis mode and property management method, reasonable spatial monitoring optimization is achieved. Different spaces in the monitoring area can be unified for real-time early warning monitoring, providing accurate and timely early warning information for all parties related to the property and improving the quality of property management.

[0061] This system realizes the collection of monitoring data for different types and different spaces through the collection units of different types of monitoring data, providing an important data basis for subsequent early warning monitoring analysis. At the same time, the analysis unit can uniformly use the historical data and real-time data collected by the collection unit for real-time monitoring analysis of the early warning state, which is efficient and accurate, and uses the early warning information processing unit to timely provide the necessary characteristic information of the early warning information to all parties related to the property. The entire system realizes real-time analysis of early warning monitoring under efficient and accurate spatial integration through simple and tight functional unit configuration, improving the quality of the system for property management and enhancing the quality of property services.

[0062] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following (items)" or its similar expressions refer to any combination of these items, including any combination of single (item) or plural items (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, c can be single or multiple.

[0063] It should be understood that in various embodiments of the present invention, the magnitude of the sequence numbers of the above processes does not mean 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.

[0064] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in connection with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled professionals 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.

[0065] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0066] In several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, 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, and there may be other division methods in actual implementation. For example, 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 displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0067] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be 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.

[0068] In addition, the functional units in the various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0069] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this 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 the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0070] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0071] The above embodiments only represent several implementation manners of the present invention, and the description thereof is relatively specific and detailed. However, it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the invention patent should be subject to the appended claims.

Claims

1. An intelligent management method based on spatial merging, characterized in that, Including: Calibrating the monitoring data types for different monitoring spaces to form a parameter-based monitoring space and an image-based monitoring space; For different types of the monitoring spaces, obtaining the historical monitoring data of the monitoring spaces, and performing feature division based on the security status on the monitoring spaces according to the historical monitoring data to form security status reference monitoring data; For different types of the monitoring spaces, obtaining the historical monitoring data of the monitoring spaces, and performing feature division based on the warning status to form warning status comparison monitoring data; Obtaining the real-time monitoring data of the monitoring spaces, and performing state-based warning monitoring analysis by combining the security status reference monitoring data and the warning status comparison monitoring data to form warning monitoring analysis result data; Performing monitoring warning according to the warning monitoring analysis result data; Among them, for different types of the monitoring spaces, obtaining the historical monitoring data of the monitoring spaces, and performing feature division based on the security status on the monitoring spaces according to the historical monitoring data to form security status reference monitoring data, including: For the parameter-based monitoring space, obtaining parameter-based historical monitoring data; Setting a parameter monitoring period, and taking the parameter monitoring period as the basis for data division, segmenting and extracting the parameter-based historical monitoring data that is not calibrated as the warning status in the parameter-based historical monitoring data in the order of the time dimension to form a periodic non-warning parameter monitoring information set A; For each parameter in the periodic non-warning parameter monitoring information set A, obtaining the numerical range on all the parameter monitoring periods, and performing an intersection operation on the numerical ranges to determine the security status range corresponding to each parameter; Aggregating the security status ranges corresponding to all the parameters in the periodic non-warning parameter monitoring information set A to form a parameter range security status monitoring set B; For the image-based monitoring space, obtaining image-based historical monitoring data; Setting an image monitoring period, and taking the image monitoring period as the basis for data division, segmenting and extracting the image-based historical monitoring data that is not calibrated as the warning status in the image-based historical monitoring data in the order of the time dimension to form a periodic non-warning image monitoring information set E; Setting a security and stability duration threshold β, and for all the images in the periodic non-warning image monitoring information set E, performing the following duration analysis and judgment on the images based on the object: If the duration of the same object existing in each image monitoring period in all the images is greater than or equal to the security and stability duration threshold β, determining the object as a secure and stable object; If the duration of the object existing in the corresponding image monitoring period in all the images is less than the security and stability duration threshold β, determining the object as an insecure and unstable object; Aggregating all the secure and stable objects to form an image object security status monitoring set F; for different types of the monitoring spaces, obtaining the historical monitoring data of the monitoring spaces, and performing feature division based on the warning status to form warning status comparison monitoring data, including: Taking the parameter monitoring period as the basis for data division, segment extraction based on the time dimension order is performed on the parameter class historical monitoring data marked with the warning status, to form a periodic warning parameter monitoring information set C; For each parameter in the periodic warning parameter monitoring information set C, obtain the value ranges on all the parameter monitoring periods, and perform a union operation on the value ranges to determine the warning status range corresponding to each parameter; Aggregate the warning status ranges corresponding to all the parameters in the periodic warning parameter monitoring information set C to form a parameter range warning status monitoring set D; Taking the image monitoring period as the basis for data division, segment extraction based on the time dimension order is performed on the image class historical monitoring data marked with the warning status, to form a periodic warning image monitoring information set G; Set a warning duration threshold η and a warning quantity threshold γ, and perform the following analysis and judgment on all the images in the periodic warning image monitoring information set G after removing the safe and stable objects in the image object safety status monitoring set F: If there exists an object whose duration in the image monitoring period is greater than or equal to the warning duration threshold η, and the number of times the object appears in different image monitoring periods is greater than or equal to the warning quantity threshold γ, then determine the object as a warning object; otherwise, determine it as a non-warning object; Aggregate all the warning objects to form an image object warning status monitoring set H; Obtain the real-time monitoring data of the monitoring space, and perform state-based warning monitoring analysis in combination with the safety status reference monitoring data and the warning status comparison monitoring data, to form warning monitoring analysis result data, including: Taking the parameter monitoring period as the time range for real-time data acquisition, periodically obtain real-time parameter monitoring data, and determine the real-time parameter value range corresponding to each parameter in the real-time parameter monitoring data; Perform the following comparative analysis on the real-time parameter value range of each parameter in the real-time parameter monitoring data and the safety status range of the corresponding parameter in the parameter range safety status monitoring set B: If the real-time parameter value ranges of all the parameters belong to the corresponding safety status ranges, then determine that the parameter class monitoring space is in a safety monitoring state; Otherwise, determine that it is in a non-safety monitoring state. For the parameter class monitoring space in the non-safety monitoring state, set a warning parameter quantity threshold α, and perform the following comparative judgment in combination with the warning status range corresponding to each parameter in the parameter range warning status monitoring set D: If the total number of parameters whose real-time parameter value ranges belong to the corresponding warning status ranges is less than the warning parameter quantity threshold α, then judge that the parameter class monitoring space is in a non-warning state; If the total number of parameters whose real-time parameter value ranges belong to the corresponding warning status ranges is greater than or equal to the warning parameter quantity threshold α, then judge that the parameter class monitoring space is in a warning state; Taking the image monitoring period as the time range for real-time data acquisition, periodically obtaining real-time image monitoring data, and determining real-time image objects existing in each image in the real-time image monitoring data; Performing the following comparative analysis on all the real-time image objects in the real-time image monitoring data and the secure and stable objects in the image object security status monitoring set F: If all the real-time image objects belong to the secure and stable objects in the image object security status monitoring set F, determining that the image class monitoring space is in a security monitoring state; Otherwise, determining that it is in a non-security monitoring state. For the image class monitoring space in the non-security monitoring state, setting an early warning duration threshold T, and combining all the early warning objects in the image object early warning status monitoring set H to perform the following comparative judgment: If there exists a real-time image object belonging to the early warning objects in the image object early warning status monitoring set H, but the duration of its existence in the corresponding image monitoring period is less than the early warning duration threshold T, determining that the image class monitoring space is in a non-early warning state; If there exists a real-time image object belonging to the early warning objects in the image object early warning status monitoring set H, but the duration of its existence in the corresponding image monitoring period is greater than or equal to the early warning duration threshold T, determining that the image class monitoring space is in an early warning state.

2. The intelligent management method based on spatial merging according to claim 1, wherein Performing monitoring and early warning according to the early warning monitoring analysis result data, including: Determining the parameter class monitoring space in the early warning state, calibrating the parameters whose real-time parameter value ranges belong to the corresponding early warning state ranges to form early warning abnormal parameters; Combining the early warning abnormal parameters and the real-time parameter value ranges corresponding to the early warning abnormal parameters to form monitoring and early warning information.

3. The intelligent management method based on spatial merging according to claim 1, characterized in that Performing monitoring and early warning according to the early warning monitoring analysis result data, including: Determining the image class monitoring space in the early warning state, calibrating the real-time image objects that belong to the early warning objects in the image object early warning status monitoring set H and whose duration of existence in the corresponding image monitoring period is greater than or equal to the early warning duration threshold T to form early warning abnormal objects; Combining the early warning abnormal objects and the duration of existence of the early warning abnormal objects to form monitoring and early warning information.

4. An intelligent management system based on spatial merging, adopting the intelligent management method based on spatial merging described in any one of claims 1-3, characterized in that, Including: A monitoring space numerical data acquisition unit for acquiring the parameter value ranges in the parameter class monitoring space to form parameter class historical monitoring data and real-time parameter monitoring data; A monitoring space image data acquisition unit for acquiring images in the image class monitoring space to form image class historical monitoring data and real-time image monitoring data; An early warning analysis unit, which is used to obtain the parameter-based historical monitoring data of the monitoring space numerical data acquisition unit and the image-based historical monitoring data of the monitoring space image data acquisition unit, form corresponding safety status reference monitoring data and warning status comparison monitoring data, obtain the real-time parameter monitoring data of the monitoring space numerical data acquisition unit and the real-time image monitoring data of the monitoring space image data acquisition unit, perform state-based early warning monitoring analysis, and form early warning monitoring analysis result data; An early warning information processing unit, which is used to obtain the early warning monitoring analysis result data formed by the early warning analysis unit, form early warning information and send it to different monitoring objects.

Images