A method for preventing falsification of personnel data using fiber optic sensing

Abstract

This invention provides a method for preventing data falsification by using fiber optic sensing, comprising: Step 1: collecting real-time video data of the gas delivery pipe and transmitting it to an edge computing node for status detection to obtain a first current state of the gas delivery pipe; Step 2: when the first current state is abnormal, collecting fiber optic data of the gas delivery pipe and transmitting it to an edge computing node for status detection to obtain abnormal data of the gas delivery pipe; Step 3: parsing the abnormal data to obtain a second current state of the gas delivery pipe, and transmitting the second current state to a cloud platform for data recording. This method, through a combination of data analysis and monitoring of gas delivery pipes, can detect suspicious signs of data falsification at gas stations, thereby achieving the goal of early detection and early handling of data falsification.

Description

Technical Field

[0001] This invention relates to the field of data falsification detection and evidence collection at air pollution source monitoring stations, and in particular to a method for preventing personnel data falsification using fiber optic sensing. Background Technology

[0002] This patent relates to the field of data falsification detection and evidence collection at air pollution source monitoring stations. Air pollution source monitoring stations refer to environmental monitoring stations placed near polluting enterprises. These stations are generally managed and maintained by the enterprises, thus posing a risk of data falsification. In practice, data falsification at these stations is common. This patent utilizes software algorithms combined with hardware and software techniques involving modifications to the data collection tubes to detect and collect evidence of data falsification. Because air quality data from air pollution source monitoring stations in the environmental protection industry involves penalties for violations and orders to cease operations, impacting enterprise profits, falsification incidents are frequent, attracting attention from the industry and society. Common falsification methods include: manually modifying detection data, damaging detection instruments, and introducing multiple gases into the air duct to interfere with data collection. Currently, there are no effective detection methods to defend against and monitor these methods.

[0003] In view of this, the present invention provides a method for preventing falsification of personnel data using fiber optic sensing. Summary of the Invention

[0004] This invention provides a method for preventing data falsification by using fiber optic sensing. This method combines data analysis and monitoring of gas conduits to detect suspicious signs of data falsification at gas stations, thereby achieving early detection and early handling of data falsification.

[0005] This invention provides a method for preventing falsification of personnel data using fiber optic sensing, comprising:

[0006] Step 1: Collect real-time video data of the air duct and transmit it to the edge computing node for state detection to obtain the first current state of the air duct;

[0007] Step 2: When the first current state is abnormal, the fiber optic data of the air duct is collected and transmitted to the edge computing node for state detection to obtain the abnormal data of the air duct.

[0008] Step 3: Analyze the abnormal data to obtain the second current state of the air duct, and transmit the second current state to the cloud platform for data recording.

[0009] In one feasible approach

[0010] Also includes:

[0011] When the air duct is determined to be ruptured based on the second current state, a reminder message is generated and played through a preset speaker to remind on-site personnel to suspend work.

[0012] In one feasible approach

[0013] Each air duct is surrounded by an optical fiber;

[0014] Each optical fiber is connected to the optical fiber transceiver analyzer.

[0015] In one feasible approach

[0016] The fiber optic transceiver analyzer is used to collect fiber optic data corresponding to each fiber and transmit the fiber optic data to the edge computing node.

[0017] In one feasible approach

[0018] The output of the edge computing node is electrically connected to the input of the cloud platform;

[0019] The output of the edge computing node is electrically connected to the input of the preset sound column.

[0020] In one feasible approach

[0021] Real-time video data of the air duct is collected and transmitted to an edge computing node for state detection to obtain the first current state of the air duct, including:

[0022] The real-time video data is transmitted to an edge computing node for decoding to obtain several image decoding data; each image encoding data is then parsed to obtain the image encoding value corresponding to each image encoding data.

[0023] Obtain the median of the image encoding values ​​for all image encoding values, and match the corresponding video brightness value to the real-time video data based on the median of the image encoding values; establish image samples with corresponding brightness based on the video brightness values ​​and a preset initial sample.

[0024] Each image decoding data is input into the image sample to generate several frames of video images;

[0025] The generation time of each video image is obtained, and the video images are sorted sequentially based on the generation time to obtain a video image sequence. A first correspondence between the video images and the sequence positions is established.

[0026] Analyze the pixel distribution corresponding to each frame of the video image sequence, and mark a pixel outline composed of several identical pixels on each frame of the video image;

[0027] Using the preset air duct contour, traverse each pixel contour and extract the pixel contour that is consistent with the preset air duct contour on each frame of video image, and record it as the target contour.

[0028] The air tube image corresponding to each target contour is acquired respectively, and a second correspondence between the video image and the air tube image is established.

[0029] Based on the first and second correspondences, an airway image sequence is established;

[0030] Obtain the differences between all adjacent frames of airway images in the airway image sequence to obtain a number of abnormal pixels;

[0031] Analyze the abnormal pixel value corresponding to each abnormal pixel;

[0032] The pixel difference between the abnormal pixel value and the standard pixel value on the airway image is calculated to obtain the abnormality level of the airway and generate the first current state of the airway.

[0033] In one feasible approach

[0034] After obtaining the first current state of the air delivery tube, the following is also included:

[0035] Analyze the first current state to obtain several abnormal areas on the air duct;

[0036] Obtain the abnormal features corresponding to each abnormal region;

[0037] If the abnormal characteristics are consistent with the preset characteristics, it is determined that the air duct has malfunctioned.

[0038] In one feasible approach

[0039] The process of collecting fiber optic data from the air duct and transmitting it to an edge computing node for status detection to obtain abnormal data of the air duct includes:

[0040] Collect fiber optic data from the air duct;

[0041] The first current state is analyzed to obtain the anomaly level of the air duct, and a corresponding data segmentation scheme is matched for the optical fiber data based on the anomaly level.

[0042] Based on the data segmentation scheme, the optical fiber data is divided into several sub-data, and each sub-data is input into the edge computing node for signal conversion to obtain the optical fiber signal corresponding to each sub-data.

[0043] Each fiber signal is input into a preset grating model for signal demodulation to obtain several demodulated signals;

[0044] Obtain the valid information contained in the demodulated signal respectively;

[0045] The valid information is input into a preset optical fiber model to obtain the corresponding optical fiber state;

[0046] Each fiber state is matched with the first current state to obtain the matching degree between each fiber state and the first current state, and then sorted according to the matching degree from large to small to obtain a similarity ranking list.

[0047] Extract the fiber optic signal corresponding to the first state similarity from the similarity ranking list, and denote it as the target signal. At the same time, obtain the sub-data corresponding to the target signal and denote it as the target sub-data. Then, obtain the first position of the target sub-data on the fiber optic data.

[0048] The target signal is sampled to obtain several sample data points, and the second position of each sample data point on the target signal is recorded during the sampling process;

[0049] Obtain the peak value of each sampled data and record the third position of each peak value on the corresponding sampled data.

[0050] Based on the first position, the second position, and the third position, and combined with the peak value of the sampled data contained in each sampled data, several optical fiber data peak values ​​contained in the optical fiber data are obtained.

[0051] Obtain the sub-fiber data corresponding to each data peak to obtain the abnormal data of the air duct.

[0052] In one feasible approach

[0053] The process of transmitting the second current state to the cloud platform for data recording includes:

[0054] The second current state is encrypted, and during the encryption process, the encrypted state is converted to a non-revision mode to obtain the encrypted state;

[0055] The encrypted state is transmitted to the cloud platform, where it is decrypted to obtain and record the abnormal data of the air duct.

[0056] In one feasible approach

[0057] Also includes:

[0058] Analyze the data type of the abnormal data to obtain the abnormal type of the air duct;

[0059] By analyzing the positional relationship of the abnormal data in the optical fiber data, the abnormal position of the air duct is obtained.

[0060] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0061] First, test video data of the air delivery tube is collected to analyze its initial current state. If the initial current state is abnormal, fiber optic data is collected for state detection to obtain the second current state of the air delivery tube, which is then transmitted to the cloud platform for recording. No human operation is required from beginning to end, fundamentally avoiding human modification. Moreover, the fiber optic cable has high accuracy; even slight abnormalities in the air delivery tube will change the fiber optic data. If other gases are injected into the air delivery tube or the air delivery tube breaks, it will be recorded in the fiber optic data, greatly improving the anti-counterfeiting capabilities.

[0062] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0063] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0064] Figure 1 This is a schematic diagram illustrating the workflow of a method for preventing falsification of personnel data using fiber optic sensing, as described in an embodiment of the present invention.

[0065] Figure 2 This is a schematic diagram of the structural composition of a method for preventing falsification of personnel data using fiber optic sensing in an embodiment of the present invention;

[0066] Figure 3 This is a schematic diagram of embodiment 5 of the method for preventing falsification of personnel data using fiber optic sensing in this invention. Detailed Implementation

[0067] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0068] Example 1

[0069] A method for preventing falsification of personnel data using fiber optic sensing, such as Figure 1 As shown, it includes:

[0070] Step 1: Collect real-time video data of the air duct and transmit it to the edge computing node for state detection to obtain the first current state of the air duct;

[0071] Step 2: When the first current state is abnormal, the fiber optic data of the air duct is collected and transmitted to the edge computing node for state detection to obtain the abnormal data of the air duct.

[0072] Step 3: Analyze the abnormal data to obtain the second current state of the air duct, and transmit the second current state to the cloud platform for data recording.

[0073] In this example, the first current state includes two states: the air tube is working normally and the air tube is working abnormally.

[0074] In this example, edge computing nodes represent nodes used for data processing;

[0075] In this example, status detection refers to the process of indirectly reflecting the status of the air duct by analyzing fiber optic data;

[0076] In this example, the fiber optic data is generated by the fiber optic cable wound around the surface of the air duct;

[0077] In this example, the abnormal data indicates that the optical fiber wrapped around the air duct is abnormally operating due to the air duct being in an abnormal working state.

[0078] In this example, the reasons for the abnormal data could be: different gases mixed in the air delivery tube, or the air delivery tube being damaged.

[0079] In this example, the second current state represents the state when the gas delivery tube malfunctions, including two states: the state of being mixed with other gases and the state of being ruptured.

[0080] In this example, transmitting the second current state to the cloud platform for data recording can achieve fully automatic data recording and avoid artificial data modification;

[0081] In this example, a pressure sensor is installed where the air tube enters the instrument.

[0082] The working principle and beneficial effects of the above technical solution are as follows: First, test video data of the air duct is collected to analyze the first current state of the air duct. When the first current state is abnormal, fiber optic data is collected for state detection to obtain the second current state of the air duct, which is then transmitted to the cloud platform for recording. No human operation is required from beginning to end, fundamentally avoiding human modification. Moreover, the fiber optic has high accuracy; even a slight abnormality in the air duct will change the fiber optic data. If other gases are injected into the air duct or the air duct is ruptured, it will be recorded in the fiber optic data, greatly improving the anti-counterfeiting capabilities.

[0083] Example 2

[0084] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing further includes:

[0085] When the air duct is confirmed to have ruptured according to the second current state, an alert message is generated and played through a preset speaker to remind on-site personnel to suspend work.

[0086] The working principle and beneficial effects of the above technical solution are as follows: By setting up a sound column, under the control of the cloud IoT platform and edge technology nodes, malicious damage to the air duct can be prevented through sound blocking, thereby reducing the probability of the air duct being damaged and improving the protection effect of the air duct.

[0087] Example 3

[0088] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing includes:

[0089] Each air duct is surrounded by an optical fiber;

[0090] Each optical fiber is connected to the optical fiber transceiver analyzer.

[0091] The working principle and beneficial effects of the above technical solution are as follows: By wrapping optical fiber around the air duct and connecting the optical fiber to the optical fiber transceiver analyzer, the foundation is laid for subsequent acquisition of optical fiber data.

[0092] Example 4

[0093] Based on Example 3, the method for preventing falsification of personnel data using fiber optic sensing is as follows:

[0094] The fiber optic transceiver analyzer is used to collect fiber optic data corresponding to each fiber and transmit the fiber optic data to the edge computing node.

[0095] The working principle and beneficial effects of the above technical solution are as follows: A specific optical signal is emitted into the optical fiber through the optical fiber analyzer, and then the optical signal is analyzed by the optical signal analyzer. This can help the staff to monitor each section of the entire air duct in real time, with a wide coverage area, avoiding the situation where traditional air ducts are difficult to monitor.

[0096] Example 5

[0097] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing, such as... Figure 3 As shown:

[0098] The output of the edge computing node is electrically connected to the input of the cloud platform;

[0099] The output of the edge computing node is electrically connected to the input of the preset sound column.

[0100] The working principle and beneficial effects of the above technical solution are as follows: By connecting the edge computing nodes to the input ends of the preset sound column and the input ends of the cloud platform, the speed of information transmission is improved and the detection process is accelerated.

[0101] Example 6

[0102] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing collects real-time video data of the air duct and transmits it to an edge computing node for state detection to obtain the first current state of the air duct, including:

[0103] The real-time video data is transmitted to an edge computing node for decoding to obtain several image decoding data; each image encoding data is then parsed to obtain the image encoding value corresponding to each image encoding data.

[0104] Obtain the median of the image encoding values ​​for all image encoding values, and match the corresponding video brightness value to the real-time video data based on the median of the image encoding values; establish image samples with corresponding brightness based on the video brightness values ​​and a preset initial sample.

[0105] Each image decoding data is input into the image sample to generate several frames of video images;

[0106] The generation time of each video image is obtained, and the video images are sorted sequentially based on the generation time to obtain a video image sequence. A first correspondence between the video images and the sequence positions is established.

[0107] Analyze the pixel distribution corresponding to each frame of the video image sequence, and mark a pixel outline composed of several identical pixels on each frame of the video image;

[0108] Using the preset air duct contour, traverse each pixel contour and extract the pixel contour that is consistent with the preset air duct contour on each frame of video image, and record it as the target contour.

[0109] The air tube image corresponding to each target contour is acquired respectively, and a second correspondence between the video image and the air tube image is established.

[0110] Based on the first and second correspondences, an airway image sequence is established;

[0111] Obtain the differences between all adjacent frames of airway images in the airway image sequence to obtain a number of abnormal pixels;

[0112] Analyze the abnormal pixel value corresponding to each abnormal pixel;

[0113] The pixel difference between the abnormal pixel value and the standard pixel value on the airway image is calculated to obtain the abnormality level of the airway and generate the first current state of the airway.

[0114] In this example, the real-time video data can originate from camera equipment;

[0115] In this example, the image decoded data represents the data used to represent a frame of an image in real-time video data;

[0116] In this example, the image encoding value represents the result after removing redundant parts from the image encoding data;

[0117] In this example, the video brightness value represents the shooting brightness of the camera device that collects real-time video data;

[0118] In this example, the preset initial sample represents a blank image sample with an initial brightness of 0;

[0119] In this example, the image sample represents the sample generated after correcting the preset initial sample brightness using the image encoding median;

[0120] In this example, the video image sequence is a sequence generated by sorting several video images according to their generation time, and can also be regarded as a real-time video stream;

[0121] In this example, the first correspondence represents the positional relationship of video images in the video image sequence;

[0122] In this example, the pixel outline represents an outline composed of pixels with the same pixel value;

[0123] In this example, the target contour refers to the pixel contour in the video image that matches the contour of the air duct.

[0124] In this example, the second correspondence represents a one-to-one correspondence between the air tube image and the video image, and the air tube image originates from the video image;

[0125] In this example, the airway image sequence represents a sequence consisting of airway images contained in each frame of video image;

[0126] In this example, the abnormal pixel indicates that the pixel value of the pixel changes when the outer surface of the air duct is ruptured.

[0127] To verify this example: Real-time video data A from the air duct is acquired and transmitted to an edge computing node for decoding, resulting in several image decoding data points a, b, c, d... These are then parsed to obtain the corresponding image encoding values ​​a1, b1, c1, d1... The median of these image encodings is denoted as z1. This median is then used to match the real-time video data with a video brightness value H. Combined with a preset initial sample, an image sample T is obtained. The image encoding values ​​a1, b1, c1, d1... are sequentially input into each image sample T to obtain several video frames Ta1, Tb1, Tc1... Td1…… is then sorted to generate a video image sequence. At this point, the position of each video image in the video image sequence is unique and corresponding. Then, the pixel contours on each frame of the video image are analyzed. Next, the preset air tube contour δ is used to traverse the pixel contours to obtain the air tube image in each video image. At this point, the video image and the air tube image are in one-to-one correspondence, and the air tube image originates from the video image. Thus, an air tube image sequence can be generated. Finally, the differences between the pixels on adjacent air tube images in the sequence are analyzed to generate the first current state of the air tube.

[0128] The working principle and beneficial effects of the above technical solution are as follows: In order to analyze whether there is any abnormality in the air tube, real-time video data is first input to the edge computing nodes for decoding to obtain image decoding data. Then, several video images are generated and analyzed to obtain the expression image of the air tube in each frame of the video image. Finally, based on the pixels of the air tube image, its abnormal state is analyzed. In this way, abnormal behavior approaching the air tube can be identified through video monitoring and uploaded to the edge technology nodes and the cloud IoT platform for easy evidence collection. Thus, the behavior of damaging the air tube can be effectively monitored remotely, and the behavior of abnormal instrument operation can also be monitored.

[0129] Example 7

[0130] Based on Example 6, the method for preventing falsification of personnel data using fiber optic sensing, after obtaining the first current state of the air duct, includes:

[0131] Analyze the first current state to obtain several abnormal areas on the air duct;

[0132] Obtain the abnormal features corresponding to each abnormal region;

[0133] If the abnormal characteristics are consistent with the preset characteristics, it is determined that the air duct has malfunctioned.

[0134] The working principle and beneficial effects of the above technical solution: Since a ruptured gas duct can lead to gas leakage and environmental pollution, analyzing whether the gas duct has ruptured when it malfunctions provides a reference for relevant personnel.

[0135] Example 8

[0136] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing, which involves collecting fiber optic data from the air duct and transmitting it to an edge computing node for status detection to obtain abnormal data of the air duct, includes:

[0137] Collect fiber optic data from the air duct;

[0138] The first current state is analyzed to obtain the anomaly level of the air duct, and a corresponding data segmentation scheme is matched for the optical fiber data based on the anomaly level.

[0139] Based on the data segmentation scheme, the optical fiber data is divided into several sub-data, and each sub-data is input into the edge computing node for signal conversion to obtain the optical fiber signal corresponding to each sub-data.

[0140] Each fiber signal is input into a preset grating model for signal demodulation to obtain several demodulated signals;

[0141] Obtain the valid information contained in the demodulated signal respectively;

[0142] The valid information is input into a preset optical fiber model to obtain the corresponding optical fiber state;

[0143] Each fiber state is matched with the first current state to obtain the matching degree between each fiber state and the first current state, and then sorted according to the matching degree from large to small to obtain a similarity ranking list.

[0144] Extract the fiber optic signal corresponding to the first state similarity from the similarity ranking list, and denote it as the target signal. At the same time, obtain the sub-data corresponding to the target signal and denote it as the target sub-data. Then, obtain the first position of the target sub-data on the fiber optic data.

[0145] The target signal is sampled to obtain several sample data points, and the second position of each sample data point on the target signal is recorded during the sampling process;

[0146] Obtain the peak value of each sampled data and record the third position of each peak value on the corresponding sampled data.

[0147] Based on the first position, the second position, and the third position, and combined with the peak value of the sampled data contained in each sampled data, several optical fiber data peak values ​​contained in the optical fiber data are obtained.

[0148] Obtain the sub-fiber data corresponding to each data peak to obtain the abnormal data of the air duct.

[0149] In this example, the segmentation scheme represents a method of dividing fiber optic data into different numbers of segments based on different anomaly levels;

[0150] In this example, the fiber optic signal can be an optical signal;

[0151] In this example, the demodulated signal represents the signal obtained after the carrier signal in the optical fiber signal is removed using a grating model;

[0152] In this example, valid information represents information that is not zero in the demodulated signal;

[0153] In this example, the preset fiber optic model represents the simulation of fiber optics in virtual space;

[0154] In this example, the fiber optic states include: normal, bent, crushed, moved, and slightly deformed.

[0155] In this example, the peak value of the sampled data represents the maximum value in the sampled data;

[0156] To verify this example: First, acquire the fiber optic data G of the air duct. Then, analyze the first current state to obtain that the abnormality level of the air duct is level two. Next, match a segmentation scheme to divide the fiber optic data G into 5 sub-data segments, and input them into the grating model to obtain 5 demodulated signals: signal 1, signal 2, signal 3, signal 4, and signal 5. Based on the effective information in these 5 signals and the preset fiber optic model, obtain 5 fiber optic states. Match these states with the first current state. The fiber optic state corresponding to signal 4 has the highest matching degree with the first current state. Signal 4 is recorded as the target signal. At the same time, record the first position of the sub-data segment corresponding to signal 4 on the fiber optic data. Then, sample the target signal to obtain 3 sampled data, namely data 1, data 2, and data 3. At the same time, record the second position of each sampled data on the target signal. Then, obtain the peak value of each sampled data and record the third position of each peak value on the sampled data. Finally, obtain the sub-fiber optic data corresponding to data 1, data 2, and data 3 on the fiber optic data to obtain the abnormal data of the air duct.

[0157] The working principle and beneficial effects of the above technical solution are as follows: Fiber optic data is demodulated to obtain several demodulated signals. Then, the effective information contained in the demodulated signals is analyzed to further determine the fiber optic state. Next, the target signal matching the first current state is extracted, sampled, and the position of the peak value of the sampled data on the fiber optic data and the corresponding sub-data are analyzed. Finally, abnormal data is generated. In this way, by analyzing the optical signal, events such as gas rupture and breakage can be detected. After the fiber optic cable breaks or breaks, the optical signal will be interrupted. By analyzing the optical signal, events such as gas rupture and breakage can be monitored, and abnormal data can be obtained.

[0158] Example 9

[0159] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing, the process of transmitting the second current state to the cloud platform for data recording, includes:

[0160] The second current state is encrypted, and during the encryption process, the encrypted state is converted to a non-revision mode to obtain the encrypted state;

[0161] The encrypted state is transmitted to the cloud platform, where it is decrypted to obtain and record the abnormal data of the air duct.

[0162] The working principle and beneficial effects of the above technical solution are as follows: In order to further prevent human tampering with data, when the second current state is transmitted to the cloud platform, it is modified to a mode that prohibits revision, thus ensuring data security.

[0163] Example 10

[0164] Based on Example 1, the method for preventing falsification of personnel data using fiber optic sensing further includes:

[0165] Analyze the data type of the abnormal data to obtain the abnormal type of the air duct;

[0166] By analyzing the positional relationship of the abnormal data in the optical fiber data, the abnormal position of the air duct is obtained.

[0167] In this example, the abnormality types include: abnormal air pressure inside the air tube and air tube rupture.

[0168] The working principle and beneficial effects of the above technical solution are as follows: By analyzing the data type of abnormal data, the abnormal type and location of the air tube can be further obtained, which facilitates timely recording.

[0169] Example 11

[0170] Based on Embodiment 1, the method for preventing falsification of personnel data using fiber optic sensing, after transmitting the second current state to the cloud platform for data recording, further includes:

[0171] Sample data is obtained by sampling the recorded data in the cloud platform at fixed real points.

[0172] Extract the last two digits of the sample data and denote them as d2d1. Use formula (1) to make a first judgment on the sample data.

[0173]

[0174] Where P represents the first judgment result, n represents the maximum value of the units digit in the last two data points of the sample data, and n = 9 under decimal adjustment, d1 represents the units digit in the last two data points of the sample data, and d2 represents the tens digit in the last two data points of the sample data.

[0175] Obtain the calculation result of formula (1) and generate the first judgment result;

[0176] Obtain the known quantities contained in the sample data, denoted as A = {X0 = i0, X1 = i1, ..., X...} n -1 = i n -1},B={X n =i n}, C = {X n +1=i n +1};

[0177] Where A represents the set of known quantities in the range [0, n-1] of the sample data, B represents the nth known quantity in the sample data, C represents the set of known quantities in the range [n+1, ∞] of the sample data, X represents the order of the known quantities in the sample data, and i represents the value corresponding to the known quantity.

[0178] The known quantity is input into a preset first-order transition matrix for a second determination;

[0179]

[0180] Obtain the calculation result of formula (2) and generate the second judgment result;

[0181] If the sample data violates both the first and second criteria, the sample will be labeled as suspected of being counterfeit.

[0182] When the sample data violates the first judgment but meets the second judgment, the sample data will be labeled as slightly suspected of being falsified.

[0183] When the sample data violates the second judgment but meets the first judgment, the sample data is labeled as abnormal.

[0184] The working principle and beneficial effects of the above technical solution are as follows: In order to further ensure data security, data in the cloud platform is sampled and judged within a preset time period to analyze whether it is abnormal, and to provide reference for relevant personnel.

[0185] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for preventing falsification of personnel data using fiber optic sensing, characterized in that, include: Step 1: Collect real-time video data of the air duct and transmit it to the edge computing node for state detection to obtain the first current state of the air duct; Step 2: When the first current state is abnormal, the fiber optic data of the air duct is collected and transmitted to the edge computing node for state detection to obtain the abnormal data of the air duct. Step 3: Analyze the abnormal data to obtain the second current state of the air duct, and transmit the second current state to the cloud platform for data recording; Real-time video data of the air duct is collected and transmitted to an edge computing node for state detection to obtain the first current state of the air duct, including: The real-time video data is transmitted to an edge computing node for decoding to obtain several image decoding data; each image encoding data is then parsed to obtain the image encoding value corresponding to each image encoding data. Obtain the median of the image encoding values ​​for all image encoding values, and match the corresponding video brightness value to the real-time video data based on the median of the image encoding values; establish image samples with corresponding brightness based on the video brightness values ​​and a preset initial sample. Each image decoding data is input into the image sample to generate several frames of video images; The generation time of each video image is obtained, and the video images are sorted sequentially based on the generation time to obtain a video image sequence. A first correspondence between the video images and the sequence positions is established. Analyze the pixel distribution corresponding to each frame of the video image sequence, and mark a pixel outline composed of several identical pixels on each frame of the video image; Using the preset air duct contour, traverse each pixel contour and extract the pixel contour that is consistent with the preset air duct contour on each frame of video image, and record it as the target contour. The air tube image corresponding to each target contour is acquired respectively, and a second correspondence between the video image and the air tube image is established. Based on the first and second correspondences, an airway image sequence is established; Obtain the differences between all adjacent frames of airway images in the airway image sequence to obtain a number of abnormal pixels; Analyze the abnormal pixel value corresponding to each abnormal pixel; Calculate the pixel difference between the abnormal pixel value and the standard pixel value on the airway image to obtain the abnormality level of the airway and generate the first current state of the airway. After transmitting the second current state to the cloud platform for data recording, the process also includes: Sample data is obtained by sampling the recorded data in the cloud platform at fixed real points. Extract the last two digits of the sample data and denote them as d2d1. Use formula (1) to make a first judgment on the sample data. Where P represents the first judgment result, n represents the maximum value of the units digit in the last two data points of the sample data, and n = 9 under decimal adjustment, d1 represents the units digit in the last two data points of the sample data, and d2 represents the tens digit in the last two data points of the sample data. Obtain the calculation result of formula (1) and generate the first judgment result; Obtain the known quantities contained in the sample data, denoted as A = {X0 = i0, X1 = i1, ..., X...} n -1 = i n -1},B={X n =i n }, C = {X n +1=i n +1}; Where A represents the set of known quantities in the range [0, n-1] of the sample data, B represents the nth known quantity in the sample data, C represents the set of known quantities in the range [n+1, ∞] of the sample data, X represents the order of the known quantities in the sample data, and i represents the value corresponding to the known quantity. The known quantity is input into a preset first-order transition matrix for a second determination; Obtain the calculation result of formula (2) and generate the second judgment result; If the sample data violates both the first and second criteria, the sample will be labeled as suspected of being counterfeit. When the sample data violates the first judgment but meets the second judgment, the sample data will be labeled as slightly suspected of being falsified. When the sample data violates the second criterion but meets the first criterion, the sample data is labeled as abnormal.

2. The method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that, Also includes: When the air duct is determined to be ruptured based on the second current state, a reminder message is generated and played through a preset speaker to remind on-site personnel to suspend work.

3. The method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that: Each air duct is surrounded by an optical fiber; Each optical fiber is connected to the optical fiber transceiver analyzer.

4. A method for preventing falsification of personnel data using fiber optic sensing as described in claim 3, characterized in that: The fiber optic transceiver analyzer is used to collect fiber optic data corresponding to each fiber and transmit the fiber optic data to the edge computing node.

5. A method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that: The output of the edge computing node is electrically connected to the input of the cloud platform; The output of the edge computing node is electrically connected to the input of the preset sound column.

6. A method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that, After obtaining the first current state of the air delivery tube, the following is also included: Analyze the first current state to obtain several abnormal areas on the air duct; Obtain the abnormal features corresponding to each abnormal region; If the abnormal characteristics are consistent with the preset characteristics, it is determined that the air duct has malfunctioned.

7. A method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that, The process of collecting fiber optic data from the air duct and transmitting it to an edge computing node for status detection to obtain abnormal data of the air duct includes: Collect fiber optic data from the air duct; The first current state is analyzed to obtain the anomaly level of the air duct, and a corresponding data segmentation scheme is matched for the optical fiber data based on the anomaly level. Based on the data segmentation scheme, the optical fiber data is divided into several sub-data, and each sub-data is input into the edge computing node for signal conversion to obtain the optical fiber signal corresponding to each sub-data. Each fiber signal is input into a preset grating model for signal demodulation to obtain several demodulated signals; The effective information contained in the demodulated signal is obtained respectively; the effective information is input into the preset optical fiber model to obtain the corresponding optical fiber state; Each fiber state is matched with the first current state to obtain the matching degree between each fiber state and the first current state, and then sorted according to the matching degree from large to small to obtain a similarity ranking list. Extract the fiber optic signal corresponding to the first state similarity from the similarity ranking list, and denote it as the target signal. At the same time, obtain the sub-data corresponding to the target signal and denote it as the target sub-data. Then, obtain the first position of the target sub-data on the fiber optic data. The target signal is sampled to obtain several sample data points, and the second position of each sample data point on the target signal is recorded during the sampling process; Obtain the peak value of each sampled data and record the third position of each peak value on the corresponding sampled data. Based on the first position, the second position, and the third position, and combined with the peak value of the sampled data contained in each sampled data, several optical fiber data peak values ​​contained in the optical fiber data are obtained. Obtain the sub-fiber data corresponding to each data peak to obtain the abnormal data of the air duct.

8. A method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that, The process of transmitting the second current state to the cloud platform for data recording includes: The second current state is encrypted, and during the encryption process, the encrypted state is converted to a non-revision mode to obtain the encrypted state; The encrypted state is transmitted to the cloud platform, where it is decrypted to obtain and record the abnormal data of the air duct.

9. A method for preventing falsification of personnel data using fiber optic sensing as described in claim 1, characterized in that, Also includes: Analyze the data type of the abnormal data to obtain the abnormal type of the air duct; By analyzing the positional relationship of the abnormal data in the optical fiber data, the abnormal position of the air duct is obtained.

Images