Construction site anti-theft early warning system based on Internet platform technology

Abstract

The invention discloses a construction site anti-theft early warning system based on the Internet platform technology, and particularly relates to the technical field of Internet anti-theft early warning, and the system comprises a personnel recognition module which collects the personnel data and object data of a target construction site; the early warning analysis module is used for analyzing the personnel data and the object data and generating an anti-theft early warning signal; the signal intelligent processing module is used for carrying out risk judgment on the anti-theft early warning signal to form a quantitative risk threshold value; the Internet cloud platform module is used for constructing a cloud storage unit, a visual management unit and a block chain evidence storage unit; and the multi-stage early warning response module is used for carrying out graded response aiming at different risk grades. Through the personnel identification module, the early warning analysis module, the signal intelligent processing module, the Internet cloud platform module and the multi-level early warning response module, the problems that the construction site anti-theft early warning system is poor in alarm accuracy, the response mechanism is not intelligent, the evidence chain credibility is insufficient and the like are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet anti-theft early warning, more particularly, the present application relates to a construction site anti-theft early warning system based on Internet platform technology. BACKGROUND

[0002] In the process of power distribution construction, a large number of cables, transformers and other power materials stored in the construction site have the characteristics of high value and easy transfer, making them the focus of thieves; in the field of anti-theft early warning technology in power distribution construction site, power distribution construction has the characteristics of temporary site, concentration of core assets (such as high-value equipment and materials such as cables, transformers, distribution boxes), complex personnel flow, and open work environment, and its anti-theft demand is becoming more and more prominent, therefore, to realize effective anti-theft, anti-invasion and anti-external damage of the construction site has become a key task to ensure the safety and efficiency of power construction.

[0003] The existing construction site anti-theft early warning system has basically met the anti-theft needs, but in actual application, it still has some shortcomings: first, the alarm logic is single, usually only for the binary mode of triggering to alarm, which cannot effectively distinguish the intention and risk level of the intrusion behavior, resulting in high false alarm rate, a large number of invalid alarm information will flood the invalid alarm, making the management personnel feel fatigue, thereby reducing the practicability and early warning effectiveness of the system; second, the response mechanism is rigid, the same alarm mode is adopted regardless of the severity of the event, which cannot realize the precise allocation of anti-theft resources; for real major theft behavior, the way of only notifying the administrator may miss the best disposal opportunity due to the personal failure to respond in time, and lacks an effective channel for connecting the theft event to public safety resources; third, the alarm data generated by the system lacks public credibility in judicial evidence level, and there is a risk of tampering, which is difficult to serve as effective legal evidence. Therefore, an anti-theft early warning system for construction site based on Internet platform technology is urgently needed to solve the problems of poor alarm accuracy, un-intelligent response mechanism, and insufficient evidence chain credibility. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides an anti-theft early warning system for construction site based on Internet platform technology, which solves the problems raised in the background art by the following scheme.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme: an anti-theft early warning system for construction site based on Internet platform technology, comprising: A personnel identification module: setting the power distribution construction site as a target construction site, deploying sensors in the target construction site, and collecting personnel data and object data of the target construction site; Early warning analysis module: connected to the personnel identification module, it analyzes personnel data and object data to generate anti-theft early warning signals, including personnel approach signals, personnel intrusion signals, and object change signals; Signal intelligent processing module: connected to the early warning analysis module, it receives anti-theft early warning signals in real time, and performs risk assessment on the anti-theft early warning signals through the built-in microcontroller to form a quantitative risk threshold. Based on the risk threshold, it generates a theft risk level, which includes level one risk, level two risk, and level three risk. Internet cloud platform module: connected to the signal intelligent processing module, it displays and stores the theft risk level based on the Internet cloud platform, which includes a cloud storage unit, a visualization management unit, and a blockchain evidence storage unit; Multi-level early warning response module: Connected to the Internet cloud platform module and the intelligent signal processing module, it will trigger an audible and visual alarm to drive away when a level 1 risk is triggered, push an administrator information alarm when a level 2 risk is triggered, and trigger a public security linkage alarm when a level 3 risk is triggered.

[0006] Preferably, the intelligent signal processing module is connected to the early warning analysis module, receives anti-theft early warning signals in real time, and performs risk assessment on the anti-theft early warning signals through a built-in microcontroller to form a quantitative risk threshold. Based on the risk threshold, a theft risk level is generated, including Level 1 risk, Level 2 risk, and Level 3 risk. A connection is established with the early warning analysis module through the UART hardware interface of the microcontroller. A custom frame format is defined, which includes: frame header, signal type, data length, content, and CRC check code. The received data frame is parsed to extract core fields, such as signal type (person approaching, intrusion, object change), associated person ID, face image D, and defense zone location. Risk assessment is performed on anti-theft warning signals to form quantitative risk thresholds; key variables for risk level assessment are parsed from data frames; and dynamic factors for signal types are constructed to assign basic risk values ​​to different signal types. When the parsed data frame is a person approaching signal, the risk value is... =30. When the parsed data frame is a personnel intrusion signal, the risk value is... =70. When the parsed data frame is an object change signal, then the risk value is... =60; Confidence based on the association between personnel data and object data A confidence factor is constructed; the confidence is calculated by superimposing the weights of the three defense zones. The confidence weight of triggering a single defense zone of type A, B, and C is 0.25; the confidence weight of triggering a combination of defense zones (A, B), (A, C), and (B, C) is 0.35; and the confidence weight of triggering all three defense zones (A, B, C) simultaneously is 0.40. Further explanation is needed for the combination of (A, B), (A, C), (B, C) defense zones: wherein the meaning of (A, B) defense zone combination is that personnel carrying material equipment leave A defense zone and enter B defense zone, and the personnel are divided into identifiable internal personnel of power distribution construction and strangers; the meaning of (A, C) defense zone combination is that personnel carrying material equipment leave A, B defense zone and enter construction boundary C defense zone; the meaning of (B, C) defense zone combination is that personnel do not touch material equipment and only enter B, C defense zone range; Further explanation is that triggering (A, B, C) three defense zones means that personnel carrying material equipment pass through B, C defense zone and escape from the power distribution construction site; Based on personnel data, the personnel theft weight is constructed When the personnel ID is consistent with the face image D of the internal personnel of the power distribution construction site, the theft weight proportion is set to 0.4; when the personnel ID is not consistent with the face image D of the internal personnel of the power distribution construction site, the theft weight proportion is set to 0.6; Based on the basic risk value , confidence , theft weight , the quantitative risk value calculation is carried out, and then the risk threshold value is obtained; the quantitative risk value calculation formula is:

[0007] Threshold value calculation example: scene 1, strangers who cannot identify the identity approach C defense zone, and the risk threshold value ; scene 2, strangers who cannot identify the identity trigger (B, C) combined defense zone, reach B defense zone and determine as intruding personnel, and the risk threshold value ; scene 3, strangers who cannot identify the identity trigger (A, B, C) three defense zones, and the risk threshold value ; The quantitative risk threshold value of the internal personnel of the power distribution construction site is calculated by the quantitative risk value calculation method of the personnel who cannot identify the identity; Based on the calculation of the risk threshold value, the risk level is determined, when the strangers approach C defense zone within 5 meters, without touching B, A defense zone, it is determined as a first-level risk; when the strangers trigger (A, B), (A, C), (B, C) combined defense zone, it is determined as a second-level risk; when the strangers trigger (A, B, C) three defense zones, it is directly determined as a third-level risk; When the identified personnel are internal personnel of the power distribution construction site, they approach C defense zone, without triggering a first-level risk; when the internal personnel trigger (A, B), (A, C) defense zone, a second-level risk is triggered; when the internal personnel trigger (A, B, C) three defense zones, a third-level risk decision is made based on the second-level risk.

[0008] The technical effects and advantages of the present application are: The application adopts a built-in quantitative risk algorithm, which comprehensively considers a signal type dynamic factor Stype, a confidence factor Cvalue based on multi-zone triggering and a personnel identity theft weight, quantitatively evaluates the intention and severity of an intrusion behavior, and improves the alarm accuracy; The application can automatically trigger a differentiated response strategy according to the calculated risk level, realizes intelligent and efficient deployment of anti-theft resources, through multi-level early warning response and signal intelligent processing, and linkage output of the risk level; The application adopts a blockchain technology in an internet cloud platform, performs hash coding and node setting on early warning signals and risk level data, and constructs an unalterable and judicially credible early warning evidence chain; The prior art data is dispersedly stored in local hosts of each construction site, and an administrator needs to log in the system one by one for checking, and cannot realize global overall planning, the application realizes global overall planning through a personnel identification module, establishes A, B and C three-level defense zones, and improves the adaptability to complex construction sites and system reliability. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 It is a system logic structure diagram of the application.

[0010] Figure 2 It is a personnel identification module structure diagram of the application.

[0011] Figure 3 It is a signal intelligent processing module structure diagram of the application.

[0012] Figure 4 It is an execution logic structure diagram of the application. DETAILED DESCRIPTION

[0013] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.

[0014] Please refer to Figures 1-4 The application provides a construction site anti-theft early warning system based on an internet platform technology, and the system is constructed through a personnel identification module, an early warning analysis module, a signal intelligent processing module, an internet cloud platform module and a multi-level early warning response module. The application discloses a construction site anti-theft early warning system based on an internet platform technology, and the system comprises: The personnel identification module sets the power distribution construction site as a target construction site, deploys sensors in the target construction site, and collects personnel data and object data of the target construction site. The early warning analysis module is connected with the personnel identification module, analyzes the personnel data and object data, generates an anti-theft early warning signal, and the anti-theft early warning signal includes a personnel approaching signal, a personnel intrusion signal, and an object change signal. The signal intelligent processing module is connected with the early warning analysis module, receives the anti-theft early warning signal in real time, judges the risk of the anti-theft early warning signal through the built-in single-chip microcomputer, forms a quantitative risk threshold, generates a theft risk level based on the risk threshold, and the risk level includes a first-level risk, a second-level risk, and a third-level risk. The Internet cloud platform module is connected with the signal intelligent processing module, displays and stores the theft risk level based on the Internet cloud platform, and the Internet cloud platform includes a cloud storage unit, a visual management unit, and a block chain storage unit. The multi-level early warning response module is connected with the Internet cloud platform module and the signal intelligent processing module, triggers a sound and light repelling alarm when a first-level risk is triggered, pushes an administrator information alarm when a second-level risk is triggered, and triggers a public security linkage alarm when a third-level risk is triggered.

[0015] The personnel identification module sets the power distribution construction site as a target construction site, deploys sensors in the target construction site, and collects personnel data and object data of the target construction site, specifically including: The personnel data includes a construction personnel ID and face image data D, and the object data includes power distribution materials X, power distribution equipment Y, material position change data , and equipment position change data ; Target construction site layout planning and zoning: The target power distribution construction site is surveyed, and the material storage area, the main equipment storage point, and the construction boundary are set as target monitoring areas; the material and equipment storage area is divided into A defense zone, the area 5 meters outside the material and equipment storage area is divided into B defense zone, and the area 5 meters outside the power distribution construction boundary is divided into C defense zone; Sensor selection and deployment planning: Based on A, B, and C defense zones, deploy sensors, deploy infrared grating sensors in C zone to form the first intrusion detection line; deploy infrared grating sensors in B zone to form the second intrusion detection line; deploy intelligent video analysis cameras and displacement sensors in A zone to form the third object intrusion detection line; Sensor installation and parameter configuration: The infrared grating sensors are installed in pairs on the fence, and the infrared transmitting end and receiving end are aligned to form multiple parallel invisible infrared detection barriers; the vibration sensor is directly fixed on the key parts of the protected equipment and materials (such as cable reel shaft, transformer shell); the intelligent video analysis camera is installed at the commanding height and supports Monitoring, ensure coverage A, B, C, and adjust the angle of view, focal length; set the response time of the infrared grating sensor ; set the vibration amplitude threshold of the vibration sensor, filter the wind blowing environmental interference, but can respond to human behavior such as knocking, carrying, etc.; set the intelligent analysis rules of the intelligent video analysis camera, such as intrusion detection in the designated area, identify alarm personnel and object data to distinguish between people, vehicles and animals to reduce false positives.

[0016] The early warning analysis module is connected with the personnel identification module to analyze personnel data and object data and generate a theft warning signal, which includes a personnel approach signal, a personnel intrusion signal, and an object change signal. The personnel data and object data are analyzed based on the single-chip central processing unit receiving raw signals from various sensors and performing standardized processing on the raw signals to reduce noise caused by electromagnetic interference. The central processing unit performs correlation analysis on signals from the same time period and the same or related areas to generate a theft warning signal, and transmits the warning signal in the form of a data frame. Further explanation is needed for the standardized processing of the raw signal, which includes filtering and normalization of the collected personnel data and object data. The filtering is a sliding average filter that takes the average of the last , equipment position change data Set a fixed window , take the average value of the last signal data as the filtering result, which is an example of the specific filtering formula:

[0017] Among them represents the material position change data filtering result, represents the material position change data from to ; The normalization processing is performed on the material position change data , equipment position change data , respectively, to establish a data set and take and , and , and the specific normalization calculation formula is and , where represents the normalization processing of material position change data to obtain ,​ representing the normalization processing of the device position change data, to obtain ; It needs to be further explained that the correlation analysis generates a theft warning signal, and the correlation data of A, B and C three protection zones are analyzed. When the intelligent video analysis camera cannot identify the personnel ID, face image data D, the infrared grating sensor distance judgment and object data change judgment are carried out: If the personnel cannot be identified within 5 meters of the power distribution construction boundary C protection zone, a personnel approach signal is generated; If the personnel cannot be identified within 5 meters of the C protection zone into the B protection zone, a personnel intrusion signal is generated; If the personnel cannot be identified to break through the B protection zone into the A protection zone to trigger the material equipment vibration sensor, an object change signal one is generated; When the intelligent video analysis camera successfully identifies the personnel ID, face image data D, directly carries out object change data judgment, if the personnel successfully triggers the A protection zone material equipment vibration sensor, an object change signal two is generated, and the signal two is reversely judged in B and A protection zones. The reverse judgment is the purpose judgment of the internal personnel carrying the material equipment breaking through B and A protection zones; The generated personnel approach signal, personnel intrusion signal and object change signal are converted into the form of data frame for warning signal transmission.

[0018] The signal intelligent processing module is connected with the warning analysis module, and receives the theft warning signal in real time. The risk of the theft warning signal is judged by the built-in single-chip microcomputer, the quantitative risk threshold is formed, the theft risk level is generated based on the risk threshold, and the risk level includes first-level risk, second-level risk and third-level risk: The single-chip microcomputer is connected with the warning analysis module through the UART hardware interface, the frame format is customized, the frame format includes frame header, signal type, data length, content and CRC check code, the received data frame is analyzed, the core field is extracted, the signal type (personnel approach, intrusion, object change), the associated personnel ID, the face image D and the protection zone position are extracted; The risk of the theft warning signal is judged, the quantitative risk threshold is formed, the key variables for risk level judgment are analyzed from the data frame, and the signal type dynamic factor is constructed, and the basic risk value is given to different signal types : When the analyzed data frame is the personnel approach signal, the risk value = 30, when the analyzed data frame is the personnel intrusion signal, the risk value = 70, and when the analyzed data frame is the object change signal, the risk value = 60; Based on the confidence degree of the correlation of personnel data and object data , the confidence factor is constructed; the confidence is calculated by superimposing three major defense zone weights, the trigger A, B, C three single defense zone confidence weight proportion is 0.25; trigger (A, B), (A, C), (B, C) defense zone combination confidence weight proportion 0.35; while triggering (A, B, C) three major defense zone confidence weight proportion 0.40; Further explanation is needed (A, B), (A, C), (B, C) defense zone combination: (A, B) defense zone combination means that personnel carrying material equipment leave A defense zone into B defense zone, the personnel are divided into identifiable power distribution construction internal personnel and strangers; (A, C) defense zone combination means that personnel carrying material equipment leave A, B defense zone into construction boundary C defense zone; (B, C) defense zone combination means that personnel do not touch material equipment, only enter B, C defense zone range; Further explanation is needed to trigger (A, B, C) three major defense zones, which means that personnel carrying material equipment pass through B, C defense zone and escape from power distribution construction site; Based on personnel data, the personnel theft weight is constructed When the personnel ID and the face image D are consistent with the internal personnel of the power distribution construction site, the theft weight proportion is set to 0.4; when the personnel ID and the face image D are not consistent with the internal personnel of the power distribution construction site, the theft weight proportion is set to 0.6; Based on the basic risk value , confidence , theft weight Quantitative risk value calculation, and then get the risk threshold value; the quantitative risk value calculation formula is:

[0019] Threshold calculation example: scene 1, there are strangers who cannot identify the identity near C defense zone, then the risk threshold value ; Scene 2, there are strangers who cannot identify the identity trigger (B, C) combination defense zone, reach B defense zone to determine as intruder, then the risk threshold value ; Scene 3, there are strangers who cannot identify the identity trigger (A, B, C) three major defense zones, then the risk threshold value ; The quantitative risk threshold value of the internal personnel of the power distribution construction site is calculated by the quantitative risk value calculation method of the personnel who cannot identify the identity; Based on the calculation of risk threshold value, the risk level is determined by the risk threshold value method, when the stranger is close to C defense zone within 5 meters, and does not touch B, A defense zone, it is determined as first level risk; When the stranger triggers (A, B), (A, C), (B, C) combination defense zone, it is determined as second level risk; When the stranger triggers (A, B, C) three major defense zones, it is directly determined as third level risk; When the identified person is an internal staff of the power distribution construction site, no first-level risk is triggered when approaching the C defense area; when the internal staff triggers (A, B) and (A, C) defense areas, a second-level risk is triggered; when the internal staff triggers (A, B, C) three defense areas, a third-level risk decision is made based on the second-level risk.

[0020] The Internet cloud platform module is connected with the signal intelligent processing module, and displays and stores the theft risk level based on the Internet cloud platform, and the Internet cloud platform includes a cloud storage unit, a visual management unit and a block chain storage unit. The core of the Internet cloud platform module is to realize the whole process management of receiving, storing, visualizing and storing the theft risk level data, which needs to connect the risk level data output by the signal intelligent processing module, and at the same time, through the cloud storage unit, the visual management unit and the block chain storage unit, the data security is guaranteed; Further explanation is needed for the reception of the theft risk level data, which deploys MQTTBroker service in the cloud platform, and takes the signal intelligent processing module as a client, adopts lightweight MQTT protocol, and uploads the risk level and related data in real time, which includes risk level, risk value, defense area position, trigger basis and associated confidence; The cloud storage unit adopts a three-layer architecture of time sequence database, relational database and risk object storage in its storage architecture design, and the storage records include personnel ID, risk level (first / second / third), A, B and C defense area position, associated evidence files (pictures, videos) and risk trigger reasons; The visual management unit pushes the newly occurring alarm events to the Web management background and mobile App in real time through cloud server pushing technology; in the management interface, it sends an alarm to the administrator in the form of pop-up window, sound and interface highlight, and the alarm information clearly shows the risk level (such as using yellow, orange and red to distinguish), position and snapshot; Global situation display: on the main dashboard of the Web management background, the defense area positions of the construction site are displayed in the form of an electronic map; and on the defense area icon, the current state is displayed by color coding: green (normal), yellow (first-level risk), orange (second-level risk) and red (third-level risk); historical data query and analysis: provide a query interface, and the administrator can retrieve historical alarm records according to time, position, risk level and other conditions; The blockchain evidence storage unit adopts a consortium blockchain, whose nodes include the construction manager, cloud platform operator, public security department, and risk level. The evidence storage data is filtered, storing only the original warning signal hash value and risk level hash value. The evidence storage process includes: risk level data hash calculation, using JSON format to calculate the SHA-256 hash value of the filtered risk levels; on-chain evidence storage transaction, where the cloud platform, as the consortium blockchain client, calls a smart contract to submit the personnel ID, risk level hash value, and evidence storage time to the blockchain network; consensus and on-chain verification, where consortium blockchain nodes verify the validity of the transaction through the PBFT consensus mechanism, and after successful verification, write the transaction into a block, generating an evidence storage number.

[0021] The multi-level early warning response module is connected to the Internet cloud platform module and the signal intelligent processing module. When a Level 1 risk is triggered, it will trigger an audible and visual deterrent alarm; when a Level 2 risk is triggered, it will trigger an administrator information push alarm; and when a Level 3 risk is triggered, it will trigger a public security linkage alarm. Receive risk level instructions from the signal intelligent processing module and synchronize them to the Internet cloud platform module to store relevant data, and then trigger the following graded response process; The Level 1 risk response method is on-site audible and visual deterrence to prevent unintentional intrusion or probing behavior by strangers, achieving on-site autonomous deterrence. Command issuance: A start command is sent to the audible and visual alarm in Zone C via wired or short-range wireless signal. Upon receiving the command, the alarm immediately executes the audible and visual deterrence, using high-brightness LED flashing and playing a piercing siren sound of at least 100 decibels. The duration of the audible and visual deterrence is 30 seconds. If the zone signal disappears during this period (personnel have been deterred), the alarm automatically stops. If the signal persists, the deterrence cycle repeats, and the system determines whether to proceed to Level 2 risk response. The level 2 risk response method is to push information to the administrator, notify the responsible personnel, and allow for manual intervention to judge and handle the situation; determine whether it is an intrusion by an unknown person or an unauthorized internal person; push level 2 risk information to the mobile terminals of one or more administrators: the push method is APP push, sending SMS messages to the administrator's dedicated APP, sending concise alarm text information to ensure that the risk level is received when the network is poor or the APP is not running; The Level 3 risk response mechanism involves a police-linked alarm system. In the event of a major theft, public safety resources are activated immediately to create legal deterrence and enable timely response. An alarm information package is generated: the cloud platform automatically prepares and formats the reported data to ensure it meets the access requirements of the public security system. The information package includes: the location of the power distribution construction site, details of the major theft of power facilities, core evidence, links to on-site video clips, and image snapshots; blockchain-based evidence storage provides the police with legal proof.

[0022] Secondly: the embodiment of the present application discloses only the structure related to the embodiment of the present application, other structures can refer to the general design, and in the case of no conflict, the same embodiment and different embodiments of the present application can be combined with each other; Finally: the above only describes the preferred embodiments of the present application and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. Construction site anti-theft early warning system based on Internet platform technology, characterized in that, The utility model relates to an anti-theft system for power distribution construction site, which comprises a personnel identification module, a pre-warning analysis module, a signal intelligent processing module, an internet cloud platform module and a multi-level pre-warning response module. The personnel identification module sets the power distribution construction site as a target construction site, deploys sensors in the target construction site, and collects personnel data and object data of the target construction site. The pre-warning analysis module is connected to the personnel identification module, analyzes the personnel data and object data, and generates an anti-theft pre-warning signal, which includes a personnel approaching signal, a personnel intrusion signal, and an object change signal. The signal intelligent processing module is connected to the pre-warning analysis module, receives the anti-theft pre-warning signal in real time, judges the risk of the anti-theft pre-warning signal through an embedded single-chip microcomputer, forms a quantitative risk threshold, generates a theft risk level based on the risk threshold, and the risk level includes a first-level risk, a second-level risk, and a third-level risk. The internet cloud platform module is connected to the signal intelligent processing module, displays and stores the theft risk level based on the internet cloud platform, and the internet cloud platform includes a cloud storage unit, a visual management unit, and a blockchain storage unit. The multi-level pre-warning response module is connected to the internet cloud platform module and the signal intelligent processing module, triggers a sound and light repelling alarm when a first-level risk is triggered, pushes an administrator information when a second-level risk is triggered, and triggers a public security linkage alarm when a third-level risk is triggered.

2. The construction site anti-theft early warning system based on Internet platform technology according to claim 1, characterized in that, The personnel data and object data of the target construction site are collected as follows: The personnel data includes a construction personnel ID and face image data D, and the object data includes power distribution materials X, power distribution equipment Y, material position change data, and equipment position change data. The personnel data and object data are analyzed to generate an anti-theft pre-warning signal as follows: The single-chip microcomputer central processing unit receives sensor raw signals, and the raw signals are standardized, including filtering and normalization. The anti-theft pre-warning signal includes a personnel approaching signal when an unknown person approaches the C defense zone, a personnel intrusion signal when an unknown person breaks through the C defense zone and enters the B defense zone, and an object change signal when an unknown person breaks through the B defense zone and enters the A defense zone and triggers a vibration sensor. 3.The construction site theft prevention and early warning system based on internet platform technology according to claim 1, characterized in that, The risk of the anti-theft pre-warning signal is judged through the embedded single-chip microcomputer to form a quantitative risk threshold. The theft risk level is generated based on the risk threshold as follows:

4. The construction site anti-theft early warning system based on Internet platform technology according to claim 1, characterized in that, When an unknown person approaches the C defense zone without touching the B and A defense zones, it is determined as a first-level risk; when an unknown person triggers (A, B), (A, C), and (B, C) defense zone combinations, it is determined as a second-level risk; and when an unknown person triggers the A, B, and C defense zones, it is directly determined as a third-level risk. The cloud storage unit, visual management unit, and blockchain storage unit of the internet cloud platform are as follows: ​ 5.The construction site theft prevention and early warning system based on internet platform technology according to claim 1, characterized in that, ​ The data frame from the early warning analysis module is received through the UART interface of the single-chip microcomputer, and the signal type, associated personnel ID, face image data and defense area position are parsed out; based on the parsing result, a signal type dynamic factor , a confidence factor and a personnel theft weight are constructed, and a quantitative risk threshold R is calculated. 6.The construction site theft prevention and early warning system based on internet platform technology according to claim 1, characterized in that, ​ ​ 7.The construction site theft prevention and early warning system based on internet platform technology of claim 1, characterized in that, ​ The cloud storage unit adopts a three-layer architecture design, and the three layers are a time series database, a relational database and a risk object storage. The stored records include a personnel ID, a risk level, a defense zone position, associated evidence files and a risk trigger cause. The visual management unit pushes alarm events to a Web management background and a mobile App in real time through cloud server pushing technology, and alarms in a pop-up window, sound and interface highlighting manner. Alarm information displays a risk level, a position and a snapshot. The blockchain storage unit stores evidence by using alliance chain technology. Alliance chain nodes include a construction party administrator, a cloud platform operator and a public security department. Stored data includes a raw early warning signal hash value and a risk level hash value. 8.The construction site theft prevention and early warning system based on internet platform technology of claim 1, characterized in that, The multi-level early warning response module specifically includes: When a first-level risk is triggered, an instruction is sent to a sound and light alarm in a C defense zone to start a high-brightness LED flash and a siren sound of no less than 100 decibels to drive away; When a second-level risk is triggered, APP pushing and short message pushing are performed on a mobile terminal of at least one administrator; When a third-level risk is triggered, a cloud platform automatically generates and formats an alarm information package, and reports the information package to a public security system through an API interface or an enterprise alarm center. The information package includes a construction site position, an event nature, a core evidence link and blockchain storage.

Images