A campus safety intelligent monitoring method and system based on intelligent Internet of Things

Abstract

The invention discloses a campus safety intelligent monitoring method and system based on intelligent Internet of Things, and relates to the technical field of Internet of Things. When the system is running, the information of the monitoring area of ​​the front-end monitoring device is acquired and processed through the front-end monitoring module and the data processing module, and integrated into a first data set and a second data set. The monitoring area abnormality detection model is established through the regional evaluation module, and after training: the front-end abnormality detection coefficient Jcxs is obtained, and matched with a preset terminal area detection evaluation threshold J, and the terminal area detection evaluation level scheme is obtained. The radiation area terminal monitoring index Fszs is obtained through the monitoring evaluation module, and then matched with a preset radiation range correction evaluation threshold F, and the abnormal terminal radiation area correction scheme is obtained. The decision module is executed, and the edge computing and intelligent processing functions of the front-end monitoring device are achieved, and early warning is achieved, and bandwidth requirements and network security are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a campus safety intelligent monitoring method and system based on intelligent Internet of Things. Background Art

[0002] Campus safety in primary and secondary schools is a widely-concerned issue in society. Campus risks are characterized by strong suddenness, great dynamism, and high uncertainty. Currently, there are many types of front-end monitoring equipment on campus. Traditional monitoring equipment lacks intelligent analysis capabilities and cannot automatically identify and warn of potential safety risks. Therefore, there are problems such as low integration of multi-source data and insufficient warning of campus safety risks. It is urgent to rely on information technology such as the Internet of Things to achieve accurate and rapid early warning monitoring of campus safety risks.

[0003] Due to the limitations of traditional security monitoring systems, building a safe campus faces many challenges. First, the monitoring and response time for security incidents is too long; second, due to the lack of monitoring blind spots and data processing capabilities, potential security threats are not discovered and handled in a timely manner. At the same time, after discovering some abnormal situations, it is not possible to know whether there are similar situations in the surrounding areas in a timely manner. It may happen that after one abnormal situation is resolved, another abnormal situation has expanded the risk. In this case, without tracing the source or checking the surrounding areas, the trend of abnormal situations may expand, causing greater losses and security risks. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a campus safety intelligent monitoring method and system based on intelligent Internet of Things, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a campus safety intelligent monitoring system based on intelligent Internet of Things, including a front-end monitoring module, a data processing module, a regional assessment module, a monitoring and assessment module and a decision-making module;

[0006] The front-end monitoring module obtains real-time monitoring image information and regional environmental information through relevant video monitors and relevant environmental sensors installed on campus to form a terminal monitoring data group;

[0007] The data processing module preprocesses the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, integrates the real-time monitoring image information after the image feature extraction processing into a first data set, and integrates the regional environmental information after the environmental information normalization processing into a second data set;

[0008] The regional assessment module uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, and obtains the front-end anomaly detection coefficient Jcxs after training and fitting, and matches it with the preset terminal area detection assessment threshold J to obtain the terminal area detection assessment level scheme;

[0009] The monitoring and evaluation module radiates the terminal monitoring equipment with abnormal conditions in a fixed area according to the terminal area detection and evaluation level scheme, obtains the real-time status of the terminal monitoring equipment in the area, and integrates it into: the radiation area terminal monitoring index Fszs, and then matches it with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction scheme;

[0010] The decision module is specifically implemented according to the terminal area detection and evaluation level scheme and the abnormal terminal radiation area correction scheme content, including the terminal area detection and evaluation level scheme content and the regulation of terminal monitoring equipment in the radiation area based on the Internet of Things.

[0011] Preferably, the front-end monitoring module includes a video image unit and a real-time environment unit;

[0012] The image unit obtains the image and video data of the monitoring area in real time through high-definition cameras or infrared cameras installed at various key locations on the campus, forming real-time monitoring image information of the terminal device;

[0013] The real-time monitoring image information includes: video image frame information after video recording cutting and coordinate information of the monitoring position;

[0014] The environmental unit collects environmental data in the area in real time through the smoke detectors, temperature sensors, humidity sensors and gas sensor equipment related sensors installed in the campus, and forms the regional environmental information of the terminal device;

[0015] The regional environmental information includes: smoke concentration, temperature information, humidity information, decibel information and gas concentration information;

[0016] The terminal monitoring data group consists of real-time monitoring image information, regional environment information, terminal device uniqueness mark and terminal device monitoring timestamp information.

[0017] Preferably, the data processing module includes a processing unit and an integration unit;

[0018] The processing unit preprocesses the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, wherein the image feature extraction processing performs noise reduction, enhancement, edge enhancement and feature extraction on the image, including extracting color information of the fire light using RGB color space or HSV color space, using inter-frame difference method or optical flow method to detect the motion area in the image, or using edge detection and contour extraction technology to identify the shape characteristics of the fire light, to obtain relevant feature information matching the fire light in the image, and the environmental information normalization processing includes minimum and maximum normalization, Z-score normalization and decimal calibration normalization;

[0019] Integrate the real-time monitoring image information after image feature extraction processing into a first data set, and integrate the regional environmental information after environmental information normalization processing into a second data set;

[0020] The first data set includes: the average value of RGB channel fluctuation Bdz, the average value of HSV channel Pjz and the average value Slz of optical flow vector;

[0021] The second data set includes: a smoke concentration value Ywz, a real-time temperature value Wdz, a real-time humidity value Sdz, and a decibel value Fbz.

[0022] Preferably, the region assessment module includes a modeling unit and a matching unit;

[0023] The modeling unit uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, performs training and analysis, and obtains: a video image frame analysis factor Fxyz and an environmental fluctuation analysis factor Hjyz, and then fits the video image frame analysis factor Fxyz and the environmental fluctuation analysis factor Hjyz obtained by analysis, and obtains: a front-end anomaly detection coefficient Jcxs;

[0024] The front-end anomaly detection coefficient Jcxs is obtained by the following calculation formula:

[0025]

[0026] Where, Jcxs represents the front-end anomaly detection coefficient, Fxyz represents the video image frame analysis factor, Hjyz represents the environmental fluctuation analysis factor, n represents the total number of video image frames, i represents the i-th video image frame, c1 and c2 represent the proportional coefficients of the video image frame analysis factor Fxyz and the environmental fluctuation analysis factor Hjyz respectively;

[0027] Wherein, 0≤c1≤1, 0≤c2≤1, and c1+c2=1, and D represents the first correction constant;

[0028] The matching unit matches the preset relevant information with the required comparison value, including matching the preset terminal area detection evaluation threshold J with the front-end abnormality detection coefficient Jcxs to obtain the terminal area detection evaluation level scheme;

[0029] The terminal area detection assessment level scheme is obtained by the following matching method:

[0030] The front-end anomaly detection coefficient Jcxs is less than the terminal area detection evaluation threshold J, and the terminal area detection evaluation result of level 1 is obtained, and the current terminal monitoring area has no anomaly evaluation plan;

[0031] The front-end abnormal detection coefficient Jcxs ≥ the terminal area detection and evaluation threshold J, obtain the terminal area detection and evaluation result of level 2, obtain the abnormal evaluation plan of the current terminal monitoring area, and last for more than three seconds, send a processing notification to the security guard on duty to handle it, execute the terminal abnormal yellow light flashing and abnormal stay away prompt voice, and send it to the monitoring and evaluation module synchronously;

[0032] When the front-end abnormal detection coefficient Jcxs ≥ twice the terminal area detection and evaluation threshold J, obtain the terminal area detection and evaluation result of level three, obtain the current terminal monitoring area early warning evaluation plan, send an early warning notification to the security guard on duty and relevant emergency contacts for processing, execute the terminal early warning red light flashing and the early warning prohibition of fixed range approaching prompt voice, and send it to the monitoring and evaluation module simultaneously.

[0033] Preferably, the video image frame analysis factor Fxyz is obtained by the following calculation formula:

[0034]

[0035] Wherein, Fxyz represents the video image frame analysis factor, Bdz represents the RGB channel fluctuation average value, Pjz represents the HSV channel average value, Slz represents the average value of the optical flow vector, f1, f2 and f3 represent the proportional coefficients of the RGB channel fluctuation average value Bdz, the HSV channel average value Pjz and the optical flow vector average value Slz respectively;

[0036] Among them, 0≤f1≤1, 0≤f2≤1, 0≤f3≤1, and f1+f2+f3=1, and K represents the second correction constant.

[0037] Preferably, the environmental fluctuation analysis factor Hjyz is obtained by the following calculation formula:

[0038]

[0039] Wherein, Hjyz represents the environmental fluctuation analysis factor, Ywz represents the smoke concentration value, Wdz represents the real-time temperature value, Sdz represents the real-time humidity value, Fbz represents the decibel value, h1, h2, h3 and h4 represent the proportional coefficients of the smoke concentration value Ywz, the real-time temperature value Wdz, the real-time humidity value Sdz and the decibel value Fbz respectively;

[0040] Among them, 0≤h1≤1, 0≤h2≤1, 0≤h3≤1, 0≤h4≤1, and h1+h2+h3+h4=1, L indicates that the third repair is normal.

[0041] Preferably, the monitoring and evaluation module includes a trigger unit and a correction unit;

[0042] The trigger unit triggers fixed-area radiation of the terminal monitoring device with abnormal conditions according to the received terminal area detection assessment level scheme content, synchronously establishes a real-time connection communication channel, obtains the real-time monitoring status of several other terminal monitoring devices in the area, and integrates them into: radiation area terminal monitoring index Fszs;

[0043] The correction unit performs matching through a preset radiation range correction evaluation threshold F to obtain a correction scheme for the abnormal terminal radiation area;

[0044] The abnormal terminal radiation area correction solution is obtained by the following matching method:

[0045] The terminal monitoring index Fszs in the radiation area is less than the correction evaluation threshold F of the radiation range. The abnormal terminal radiation area has no abnormal evaluation plan, and the terminal area detection evaluation level plan content is not modified.

[0046] The terminal monitoring index Fszs in the radiation area is ≥ the correction assessment threshold F of the radiation range. The abnormal assessment plan within the radiation area of ​​the abnormal terminal is obtained, and the content of the terminal area detection and assessment level plan is revised, including synchronously sending abnormal terminals in the radiation area to relevant personnel for verification and disposal within a fixed time and expanding the relevant range prompts to 120% for notification.

[0047] Preferably, the radiation area terminal monitoring index Fszs is obtained by the following calculation formula:

[0048]

[0049] Where Fszs represents the terminal monitoring index in the radiation area, k represents the total number of terminal monitoring devices in the radiation area, j represents the jth terminal monitoring device in the radiation area, and w j It represents the ratio of the distance between the jth terminal monitoring device and the abnormal terminal monitoring device in the radiation area and the radiation radius. The formula is used for calculation, where Fsbj represents the radiation radius value, Zdjl represents the distance value between the abnormal terminal monitoring device and the current terminal monitoring device, and Jcxs represents the front-end abnormality detection coefficient.

[0050] Preferably, the decision module includes an execution unit;

[0051] The execution unit performs specific execution according to the terminal area detection and evaluation level plan and the abnormal terminal radiation area correction plan, including the terminal area detection and evaluation level plan content and terminal monitoring equipment positioning information to relevant personnel and lighting and voice control of terminal monitoring equipment in the radiation area based on the Internet of Things.

[0052] A campus safety intelligent monitoring method based on intelligent Internet of Things includes the following steps:

[0053] Step 1: The front-end monitoring module obtains real-time monitoring image information and regional environmental information through relevant video monitors and relevant environmental sensors installed on campus to form a terminal monitoring data group;

[0054] Step 2: The data processing module pre-processes the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, integrates the real-time monitoring image information after the image feature extraction processing into a first data set, and integrates the regional environmental information after the environmental information normalization processing into a second data set;

[0055] Step 3: The regional assessment module uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, and obtains the front-end anomaly detection coefficient Jcxs after training and fitting, and matches it with the preset terminal area detection assessment threshold J to obtain the terminal area detection assessment level scheme;

[0056] Step 4: The monitoring and evaluation module radiates the terminal monitoring equipment with abnormal conditions in a fixed area according to the terminal area detection and evaluation level plan, obtains the real-time status of the terminal monitoring equipment in the area, and integrates it into: the radiation area terminal monitoring index Fszs, and then matches it with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction plan;

[0057] Step 5: The policy module is implemented according to the terminal area detection and evaluation level plan and the abnormal terminal radiation area correction plan, including the terminal area detection and evaluation level plan and the regulation of terminal monitoring equipment in the radiation area based on the Internet of Things.

[0058] The present invention provides a campus safety intelligent monitoring method and system based on intelligent Internet of Things, which has the following beneficial effects:

[0059] (1) When the system is running, the information of the monitoring area of ​​the front-end monitoring equipment is acquired and processed through the front-end monitoring module and the data processing module, and integrated into the first data set and the second data set. The monitoring area anomaly detection model is established through the regional evaluation module. After training, the front-end anomaly detection coefficient Jcxs is obtained, and it is matched with the preset terminal area detection evaluation threshold J to obtain the terminal area detection evaluation level scheme. The radiation area terminal monitoring index Fszs is obtained through the monitoring evaluation module, and then matched with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction scheme. The decision module is executed to achieve the edge computing and intelligent processing functions of the front-end monitoring equipment, and to reduce the delay after the risk event occurs as much as possible to cause a greater safety accident, so as to achieve early warning, reduce bandwidth requirements and network security issues.

[0060] (2) Through the calculation of the video image frame analysis factor Fxyz, the environmental fluctuation analysis factor Hjyz and the front-end anomaly detection coefficient Jcxs and the acquisition of the terminal area detection assessment level scheme, a multi-level security assessment and response mechanism is formed to ensure that potential security threats can be discovered and handled in a timely manner, providing more comprehensive security monitoring.

[0061] (3) By calculating the radiation area terminal monitoring index Fszs and obtaining the abnormal terminal radiation area correction plan, the problem that the traditional system cannot promptly understand whether there are similar situations in the surrounding area after discovering an abnormality is solved, preventing the trend of abnormal situations from expanding and reducing greater losses and safety risks.

[0062] (4) Through steps 1 to 5, the information in the monitoring area of ​​the terminal monitoring equipment is accurately processed and calculated to obtain: the front-end abnormal detection coefficient Jcxs and the radiation area terminal monitoring index Fszs, and the terminal area detection evaluation level plan and the abnormal terminal radiation area correction plan are implemented and corrected simultaneously to provide effective treatment measures for the abnormal monitoring area. After detecting an abnormal situation, the system can monitor the radiation in a fixed area and obtain the real-time status of the surrounding terminal equipment to ensure full coverage of high-risk areas. This solves the problem that traditional manual patrols are time-consuming and labor-intensive and cannot achieve all-weather and full coverage monitoring, as well as ensure full coverage and comprehensive evaluation of high-risk areas, and timely discover and deal with potential security threats. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a schematic diagram of a block diagram of a campus safety intelligent monitoring system based on intelligent Internet of Things in the present invention;

[0064] Figure 2 The present invention is a schematic diagram of the steps of a campus safety intelligent monitoring method based on intelligent Internet of Things. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0066] Example 1

[0067] The present invention provides a campus safety intelligent monitoring system based on intelligent Internet of Things. Figure 1 , including front-end monitoring module, data processing module, regional assessment module, monitoring and evaluation module and decision-making module;

[0068] The front-end monitoring module obtains real-time monitoring image information and regional environmental information through relevant video monitors and relevant environmental sensors installed on campus to form a terminal monitoring data group;

[0069] The data processing module preprocesses the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, integrates the real-time monitoring image information after the image feature extraction processing into a first data set, and integrates the regional environmental information after the environmental information normalization processing into a second data set;

[0070] The regional assessment module uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, and obtains the front-end anomaly detection coefficient Jcxs after training and fitting, and matches it with the preset terminal area detection assessment threshold J to obtain the terminal area detection assessment level scheme;

[0071] The monitoring and evaluation module radiates the terminal monitoring equipment with abnormal conditions in a fixed area according to the terminal area detection and evaluation level scheme, obtains the real-time status of the terminal monitoring equipment in the area, and integrates it into: the radiation area terminal monitoring index Fszs, and then matches it with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction scheme;

[0072] The decision module is specifically implemented according to the terminal area detection and evaluation level scheme and the abnormal terminal radiation area correction scheme content, including the terminal area detection and evaluation level scheme content and the regulation of terminal monitoring equipment in the radiation area based on the Internet of Things.

[0073] In this embodiment, the information of the monitoring area of ​​the front-end monitoring device is acquired and processed through the front-end monitoring module and the data processing module, and integrated into the first data set and the second data set. The monitoring area anomaly detection model is established through the regional evaluation module. After training, the front-end anomaly detection coefficient Jcxs is obtained, and it is matched with the preset terminal area detection evaluation threshold J to obtain the terminal area detection evaluation level scheme. The radiation area terminal monitoring index Fszs is obtained through the monitoring and evaluation module, and then matched with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction scheme, which is executed through the decision-making module, achieving the edge computing and intelligent processing functions of the front-end monitoring device, reducing the delay after the risk event occurs as much as possible to cause a greater safety accident, achieving early warning, and reducing bandwidth requirements and network security issues.

[0074] Example 2

[0075] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the front-end monitoring module includes a video image unit and a real-time environment unit;

[0076] The image unit obtains the image and video data of the monitoring area in real time through high-definition cameras or infrared cameras installed at various key locations on the campus, forming real-time monitoring image information of the terminal device;

[0077] The real-time monitoring image information includes: video image frame information after video recording cutting and coordinate information of the monitoring position;

[0078] The environmental unit collects environmental data in the area in real time through the smoke detectors, temperature sensors, humidity sensors and gas sensor equipment related sensors installed in the campus, and forms the regional environmental information of the terminal device;

[0079] The regional environmental information includes: smoke concentration, temperature information, humidity information, decibel information and gas concentration information;

[0080] The terminal monitoring data group consists of real-time monitoring image information, regional environment information, terminal device uniqueness mark and terminal device monitoring timestamp information.

[0081] The data processing module includes a processing unit and an integration unit;

[0082] The processing unit preprocesses the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, wherein the image feature extraction processing performs noise reduction, enhancement, edge enhancement and feature extraction on the image, including extracting color information of the fire light using RGB color space or HSV color space, using inter-frame difference method or optical flow method to detect the motion area in the image, or using edge detection and contour extraction technology to identify the shape characteristics of the fire light, to obtain relevant feature information matching the fire light in the image, and the environmental information normalization processing includes minimum and maximum normalization, Z-score normalization and decimal calibration normalization;

[0083] Integrate the real-time monitoring image information after image feature extraction processing into a first data set, and integrate the regional environmental information after environmental information normalization processing into a second data set;

[0084] The first data set includes: the average value of RGB channel fluctuation Bdz, the average value of HSV channel Pjz and the average value Slz of optical flow vector;

[0085] The second data set includes: a smoke concentration value Ywz, a real-time temperature value Wdz, a real-time humidity value Sdz, and a decibel value Fbz.

[0086] Example 3

[0087] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the regional assessment module includes a modeling unit and a matching unit;

[0088] The modeling unit uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, performs training and analysis, and obtains: a video image frame analysis factor Fxyz and an environmental fluctuation analysis factor Hjyz, and then fits the video image frame analysis factor Fxyz and the environmental fluctuation analysis factor Hjyz obtained by analysis, and obtains: a front-end anomaly detection coefficient Jcxs;

[0089] The front-end anomaly detection coefficient Jcxs is obtained by the following calculation formula:

[0090]

[0091] Where, Jcxs represents the front-end anomaly detection coefficient, Fxyz represents the video image frame analysis factor, Hjyz represents the environmental fluctuation analysis factor, n represents the total number of video image frames, i represents the i-th video image frame, c1 and c2 represent the proportional coefficients of the video image frame analysis factor Fxyz and the environmental fluctuation analysis factor Hjyz respectively;

[0092] Wherein, 0≤c1≤1, 0≤c2≤1, and c1+c2=1, and D represents the first correction constant;

[0093] The matching unit matches the preset relevant information with the required comparison value, including matching the preset terminal area detection evaluation threshold J with the front-end abnormality detection coefficient Jcxs to obtain the terminal area detection evaluation level scheme;

[0094] The terminal area detection assessment level scheme is obtained by the following matching method:

[0095] The front-end anomaly detection coefficient Jcxs is less than the terminal area detection evaluation threshold J, and the terminal area detection evaluation result of level 1 is obtained, and the current terminal monitoring area has no anomaly evaluation plan;

[0096] The front-end abnormal detection coefficient Jcxs ≥ the terminal area detection and evaluation threshold J, obtain the terminal area detection and evaluation result of level 2, obtain the abnormal evaluation plan of the current terminal monitoring area, and last for more than three seconds, send a processing notification to the security guard on duty to handle it, execute the terminal abnormal yellow light flashing and abnormal stay away prompt voice, and send it to the monitoring and evaluation module synchronously;

[0097] When the front-end abnormal detection coefficient Jcxs ≥ twice the terminal area detection and evaluation threshold J, obtain the terminal area detection and evaluation result of level three, obtain the current terminal monitoring area early warning evaluation plan, send an early warning notification to the security guard on duty and relevant emergency contacts for processing, execute the terminal early warning red light flashing and the early warning prohibition of fixed range approaching prompt voice, and send it to the monitoring and evaluation module simultaneously.

[0098] The video image frame analysis factor Fxyz is obtained by the following calculation formula:

[0099]

[0100] Wherein, Fxyz represents the video image frame analysis factor, Bdz represents the RGB channel fluctuation average value, Pjz represents the HSV channel average value, Slz represents the average value of the optical flow vector, f1, f2 and f3 represent the proportional coefficients of the RGB channel fluctuation average value Bdz, the HSV channel average value Pjz and the optical flow vector average value Slz respectively;

[0101] Among them, 0≤f1≤1, 0≤f2≤1, 0≤f3≤1, and f1+f2+f3=1, and K represents the second correction constant.

[0102] The environmental fluctuation analysis factor Hjyz is obtained by the following calculation formula:

[0103]

[0104] Wherein, Hjyz represents the environmental fluctuation analysis factor, Ywz represents the smoke concentration value, Wdz represents the real-time temperature value, Sdz represents the real-time humidity value, Fbz represents the decibel value, h1, h2, h3 and h4 represent the proportional coefficients of the smoke concentration value Ywz, the real-time temperature value Wdz, the real-time humidity value Sdz and the decibel value Fbz respectively;

[0105] Among them, 0≤h1≤1, 0≤h2≤1, 0≤h3≤1, 0≤h4≤1, and h1+h2+h3+h4=1, L indicates that the third repair is normal.

[0106] In this embodiment, a multi-level security assessment and response mechanism is formed by calculating the video image frame analysis factor Fxyz, the environmental fluctuation analysis factor Hjyz and the front-end anomaly detection coefficient Jcxs and obtaining the terminal area detection assessment level scheme, ensuring that potential security threats can be discovered and handled in a timely manner, providing more comprehensive security monitoring.

[0107] Example 4

[0108] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the monitoring and evaluation module includes a trigger unit and a correction unit;

[0109] The trigger unit triggers fixed-area radiation of the terminal monitoring device with abnormal conditions according to the received terminal area detection assessment level scheme content, synchronously establishes a real-time connection communication channel, obtains the real-time monitoring status of several other terminal monitoring devices in the area, and integrates them into: radiation area terminal monitoring index Fszs;

[0110] The correction unit performs matching through a preset radiation range correction evaluation threshold F to obtain a correction scheme for the abnormal terminal radiation area;

[0111] The abnormal terminal radiation area correction solution is obtained by the following matching method:

[0112] The terminal monitoring index Fszs in the radiation area is less than the correction evaluation threshold F of the radiation range. The abnormal terminal radiation area has no abnormal evaluation plan, and the terminal area detection evaluation level plan content is not modified.

[0113] The terminal monitoring index Fszs in the radiation area is ≥ the correction assessment threshold F of the radiation range. The abnormal assessment plan within the radiation area of ​​the abnormal terminal is obtained, and the content of the terminal area detection and assessment level plan is revised, including synchronously sending abnormal terminals in the radiation area to relevant personnel for verification and disposal within a fixed time and expanding the relevant range prompts to 120% for notification.

[0114] The radiation area terminal monitoring index Fszs is obtained by the following calculation formula:

[0115]

[0116] Where Fszs represents the terminal monitoring index in the radiation area, k represents the total number of terminal monitoring devices in the radiation area, j represents the jth terminal monitoring device in the radiation area, and w j It represents the ratio of the distance between the jth terminal monitoring device and the abnormal terminal monitoring device in the radiation area and the radiation radius. The formula is used for calculation, where Fsbj represents the radiation radius value, Zdjl represents the distance value between the abnormal terminal monitoring device and the current terminal monitoring device, and Jcxs represents the front-end abnormality detection coefficient.

[0117] The decision module includes an execution unit;

[0118] The execution unit performs specific execution according to the terminal area detection and evaluation level plan and the abnormal terminal radiation area correction plan, including the terminal area detection and evaluation level plan content and terminal monitoring equipment positioning information to relevant personnel and lighting and voice control of terminal monitoring equipment in the radiation area based on the Internet of Things.

[0119] In this embodiment, by calculating the radiation area terminal monitoring index Fszs and obtaining the abnormal terminal radiation area correction plan, the problem that the traditional system cannot promptly understand whether there are similar situations in the surrounding areas after discovering the anomaly is solved, preventing the trend of the abnormal situation from expanding and reducing greater losses and safety risks.

[0120] Example 5

[0121] A campus safety intelligent monitoring method based on intelligent Internet of Things, please refer to Figure 2 , specifically: including the following steps:

[0122] Step 1: The front-end monitoring module obtains real-time monitoring image information and regional environmental information through relevant video monitors and relevant environmental sensors installed on campus to form a terminal monitoring data group;

[0123] Step 2: The data processing module pre-processes the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, integrates the real-time monitoring image information after the image feature extraction processing into a first data set, and integrates the regional environmental information after the environmental information normalization processing into a second data set;

[0124] Step 3: The regional assessment module uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, and obtains the front-end anomaly detection coefficient Jcxs after training and fitting, and matches it with the preset terminal area detection assessment threshold J to obtain the terminal area detection assessment level scheme;

[0125] Step 4: The monitoring and evaluation module radiates the terminal monitoring equipment with abnormal conditions in a fixed area according to the terminal area detection and evaluation level plan, obtains the real-time status of the terminal monitoring equipment in the area, and integrates it into: the radiation area terminal monitoring index Fszs, and then matches it with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction plan;

[0126] Step 5: The policy module is implemented according to the terminal area detection and evaluation level plan and the abnormal terminal radiation area correction plan, including the terminal area detection and evaluation level plan and the regulation of terminal monitoring equipment in the radiation area based on the Internet of Things.

[0127] In this embodiment, the information in the monitoring area of ​​the terminal monitoring device is accurately processed and calculated through steps one to five to obtain: the front-end abnormal detection coefficient Jcxs and the radiation area terminal monitoring index Fszs, and the terminal area detection evaluation level scheme and the abnormal terminal radiation area correction scheme are executed and corrected synchronously to provide effective processing measures for the monitored abnormal area. After detecting the abnormal situation, the system can monitor the radiation in the fixed area and obtain the real-time status of the surrounding terminal devices to ensure full coverage of high-risk areas, which solves the problem that traditional manual patrols are time-consuming and labor-intensive and cannot achieve all-weather and full coverage monitoring, as well as ensure full coverage and comprehensive evaluation of high-risk areas, and timely discover and deal with potential security threats.

[0128] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A campus safety intelligent monitoring system based on intelligent Internet of Things, characterized by: It includes front-end monitoring module, data processing module, regional assessment module, monitoring and assessment module and decision-making module; The front-end monitoring module obtains real-time monitoring image information and regional environmental information through relevant video monitors and relevant environmental sensors installed on campus to form a terminal monitoring data group; The data processing module preprocesses the terminal monitoring data group acquired by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, integrates the real-time monitoring image information after the image feature extraction processing into a first data set, and integrates the regional environmental information after the environmental information normalization processing into a second data set; The regional assessment module uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, and obtains the front-end anomaly detection coefficient Jcxs after training and fitting, and matches it with the preset terminal area detection assessment threshold J to obtain the terminal area detection assessment level scheme; The regional assessment module includes a modeling unit and a matching unit; The modeling unit uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, performs training and analysis, and obtains: a video image frame analysis factor Fxyz and an environmental fluctuation analysis factor Hjyz, and then fits the video image frame analysis factor Fxyz and the environmental fluctuation analysis factor Hjyz obtained by analysis, and obtains: a front-end anomaly detection coefficient Jcxs; The front-end anomaly detection coefficient Jcxs is obtained by the following calculation formula: Where, Jcxs represents the front-end anomaly detection coefficient, Fxyz represents the video image frame analysis factor, Hjyz represents the environmental fluctuation analysis factor, n represents the total number of video image frames, i represents the i-th video image frame, c1 and c2 represent the proportional coefficients of the video image frame analysis factor Fxyz and the environmental fluctuation analysis factor Hjyz respectively; Wherein, 0≤c1≤1, 0≤c2≤1, and c1+c2=1, and D represents the first correction constant; The matching unit matches the preset relevant information with the required comparison value, including matching the preset terminal area detection evaluation threshold J with the front-end abnormality detection coefficient Jcxs to obtain the terminal area detection evaluation level scheme; The terminal area detection assessment level scheme is obtained by the following matching method: The front-end anomaly detection coefficient Jcxs is less than the terminal area detection evaluation threshold J, and the terminal area detection evaluation result of level 1 is obtained, and the current terminal monitoring area has no anomaly evaluation plan; The front-end abnormal detection coefficient Jcxs ≥ the terminal area detection and evaluation threshold J, obtain the terminal area detection and evaluation result of level 2, obtain the abnormal evaluation plan of the current terminal monitoring area, and last for more than three seconds, send a processing notification to the security guard on duty to handle it, execute the terminal abnormal yellow light flashing and abnormal stay away prompt voice, and send it to the monitoring and evaluation module synchronously; When the front-end abnormal detection coefficient Jcxs ≥ twice the terminal area detection and evaluation threshold J, obtain the terminal area detection and evaluation result of level three, obtain the current terminal monitoring area early warning evaluation plan, send an early warning notification to the security guard on duty and the relevant emergency contacts to handle it, execute the terminal early warning red light flashing and the early warning prohibition fixed range approach prompt voice, and send it to the monitoring and evaluation module simultaneously; The video image frame analysis factor Fxyz is obtained by the following calculation formula: Wherein, Fxyz represents the video image frame analysis factor, Bdz represents the RGB channel fluctuation average value, Pjz represents the HSV channel average value, Slz represents the average value of the optical flow vector, f1, f2 and f3 represent the proportional coefficients of the RGB channel fluctuation average value Bdz, the HSV channel average value Pjz and the optical flow vector average value Slz respectively; Wherein, 0≤f1≤1, 0≤f2≤1, 0≤f3≤1, and f1+f2+f3=1, K represents the second correction constant; The environmental fluctuation analysis factor Hjyz is obtained by the following calculation formula: Wherein, Hjyz represents the environmental fluctuation analysis factor, Ywz represents the smoke concentration value, Wdz represents the real-time temperature value, Sdz represents the real-time humidity value, Fbz represents the decibel value, h1, h2, h3 and h4 represent the proportional coefficients of the smoke concentration value Ywz, the real-time temperature value Wdz, the real-time humidity value Sdz and the decibel value Fbz respectively; Among them, 0≤h1≤1, 0≤h2≤1, 0≤h3≤1, 0≤h4≤1, and h1+h2+h3+h4=1, L means the third repair is normal; The monitoring and evaluation module radiates the terminal monitoring equipment with abnormal conditions in a fixed area according to the terminal area detection and evaluation level scheme, obtains the real-time status of the terminal monitoring equipment in the area, and integrates it into: the radiation area terminal monitoring index Fszs, and then matches it with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction scheme; The monitoring and evaluation module includes a trigger unit and a correction unit; The trigger unit triggers fixed-area radiation of the terminal monitoring device with abnormal conditions according to the received terminal area detection assessment level scheme content, synchronously establishes a real-time connection communication channel, obtains the real-time monitoring status of several other terminal monitoring devices in the area, and integrates them into: radiation area terminal monitoring index Fszs; The correction unit performs matching through a preset radiation range correction evaluation threshold F to obtain a correction scheme for the abnormal terminal radiation area; The abnormal terminal radiation area correction solution is obtained by the following matching method: The terminal monitoring index Fszs in the radiation area is less than the correction evaluation threshold F of the radiation range. The abnormal terminal radiation area has no abnormal evaluation plan, and the terminal area detection evaluation level plan content is not modified. The terminal monitoring index Fszs in the radiation area is ≥ the correction assessment threshold F of the radiation range. The abnormal assessment plan in the radiation area of ​​the abnormal terminal is obtained, and the content of the terminal area detection and assessment level plan is revised, including synchronously sending abnormal terminals in the radiation area to relevant personnel for verification and disposal within a fixed time and expanding the relevant range prompts to 120% for notification; The radiation area terminal monitoring index Fszs is obtained by the following calculation formula: Where Fszs represents the terminal monitoring index in the radiation area, k represents the total number of terminal monitoring devices in the radiation area, j represents the jth terminal monitoring device in the radiation area, and w j It represents the ratio of the distance between the jth terminal monitoring device and the abnormal terminal monitoring device in the radiation area and the radiation radius. The formula is used for calculation, where Fsbj represents the radiation radius value, Zdjl represents the distance between the abnormal terminal monitoring device and the current terminal monitoring device, and Jcxs represents the front-end abnormal detection coefficient; The decision module is specifically implemented according to the terminal area detection and evaluation level scheme and the abnormal terminal radiation area correction scheme content, including the terminal area detection and evaluation level scheme content and the regulation of terminal monitoring equipment in the radiation area based on the Internet of Things.

2. According to claim 1, a campus safety intelligent monitoring system based on intelligent Internet of Things is characterized by: The front-end monitoring module includes a video image unit and a real-time environment unit; The image unit obtains the image and video data of the monitoring area in real time through high-definition cameras or infrared cameras installed at various key locations on the campus, forming real-time monitoring image information of the terminal device; The real-time monitoring image information includes: video image frame information after video recording cutting and coordinate information of the monitoring position; The environmental unit collects environmental data in the area in real time through the smoke detectors, temperature sensors, humidity sensors and gas sensor equipment related sensors installed in the campus, and forms the regional environmental information of the terminal device; The regional environmental information includes: smoke concentration, temperature information, humidity information, decibel information and gas concentration information; The terminal monitoring data group consists of real-time monitoring image information, regional environment information, terminal device uniqueness mark and terminal device monitoring timestamp information.

3. According to claim 1, a campus safety intelligent monitoring system based on intelligent Internet of Things is characterized by: The data processing module includes a processing unit and an integration unit; The processing unit preprocesses the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, wherein the image feature extraction processing performs noise reduction, enhancement, edge enhancement and feature extraction on the image, including extracting color information of the fire light using RGB color space or HSV color space, using inter-frame difference method or optical flow method to detect the motion area in the image, or using edge detection and contour extraction technology to identify the shape characteristics of the fire light, to obtain relevant feature information matching the fire light in the image, and the environmental information normalization processing includes minimum and maximum normalization, Z-score normalization and decimal calibration normalization; Integrate the real-time monitoring image information after image feature extraction processing into a first data set, and integrate the regional environmental information after environmental information normalization processing into a second data set; The first data set includes: the average value of RGB channel fluctuation Bdz, the average value of HSV channel Pjz and the average value Slz of optical flow vector; The second data set includes: a smoke concentration value Ywz, a real-time temperature value Wdz, a real-time humidity value Sdz, and a decibel value Fbz.

4. According to claim 1, a campus safety intelligent monitoring system based on intelligent Internet of Things is characterized by: The decision module includes an execution unit; The execution unit performs specific execution according to the terminal area detection and evaluation level plan and the abnormal terminal radiation area correction plan, including the terminal area detection and evaluation level plan content and terminal monitoring equipment positioning information to relevant personnel and lighting and voice control of terminal monitoring equipment in the radiation area based on the Internet of Things.

5. A campus safety intelligent monitoring method based on intelligent Internet of Things, comprising a campus safety intelligent monitoring system based on intelligent Internet of Things as described in any one of claims 1 to 4, characterized in that: The following steps are involved: Step 1: The front-end monitoring module obtains real-time monitoring image information and regional environmental information through relevant video monitors and relevant environmental sensors installed on campus to form a terminal monitoring data group; Step 2: The data processing module pre-processes the terminal monitoring data group obtained by the terminal monitoring device, including image feature extraction processing and environmental information normalization processing, integrates the real-time monitoring image information after the image feature extraction processing into a first data set, and integrates the regional environmental information after the environmental information normalization processing into a second data set; Step 3: The regional assessment module uses deep learning technology to establish a monitoring area anomaly detection model for the first data set and the second data set, and obtains the front-end anomaly detection coefficient Jcxs after training and fitting, and matches it with the preset terminal area detection assessment threshold J to obtain the terminal area detection assessment level scheme; Step 4: The monitoring and evaluation module radiates the terminal monitoring equipment with abnormal conditions in a fixed area according to the terminal area detection and evaluation level plan, obtains the real-time status of the terminal monitoring equipment in the area, and integrates it into: the radiation area terminal monitoring index Fszs, and then matches it with the preset radiation range correction evaluation threshold F to obtain the abnormal terminal radiation area correction plan; Step 5: The policy module is implemented according to the terminal area detection and evaluation level plan and the abnormal terminal radiation area correction plan, including the terminal area detection and evaluation level plan and the regulation of terminal monitoring equipment in the radiation area based on the Internet of Things.

Images