A method for detecting correct seatbelt wearing

Abstract

This invention discloses a method for detecting correct seatbelt wearing, including a data acquisition device installed on the waist of the seatbelt, with preset interval thresholds within the data acquisition device. The data acquisition steps of the data acquisition device are as follows: S1: After the pressure sensor is powered on, it continuously collects pressure values ​​and pre-stores the pressure values ​​in a real-time dynamic data acquisition device; S2: The real-time dynamic data acquisition device compares the collected pressure values ​​with the preset interval thresholds; S3: If pressure values ​​exceeding the interval threshold range are continuous, the program determines that an alarm has been triggered. Based on this, it is possible to determine whether the seatbelt is worn correctly, and thus promptly remind the user to fasten the seatbelt securely, thereby reducing the occurrence of safety accidents. In particular, it facilitates the collection of data from the user's waist, and through data analysis, timely and accurate reminders can be given to the user to wear the seatbelt correctly.

Description

Technical Field

[0001] This invention relates to the field of seat belt technology, specifically to a method for detecting the correct wearing of a seat belt. Background Technology

[0002] Geological disaster prevention and slope protection projects often use anti-slide piles for support, which requires narrow and deep foundation pit excavation. During narrow foundation pit excavation, the safety of workers is always a key focus of project safety management. Intelligent safety control equipment for narrow foundation pit excavation operations provides a guarantee for project safety management. Improperly worn safety belts can easily lead to accidents; in particular, existing detection equipment only has a fixed preset value. If the collected data exceeds or falls below the preset value, the safety belt immediately alarms, increasing the probability of false alarms and easily causing user resentment or fright. Summary of the Invention

[0003] The purpose of this invention is to provide a method for detecting the correct wearing of seat belts, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting the correct wearing of a seat belt, characterized by the following steps: S1: after the pressure sensor located at the waist of the seat belt is powered on, it will continuously collect the pressure value of the seat belt and store the pressure value in the real-time dynamic data acquisition device of the sensor;

[0005] S2: The real-time pressure data acquired by the sensor data acquisition device is used to construct a sample variance of the pressure values. ;

[0006] S3: The real-time dynamic acquisition device for sensor data will construct the sample variance. Compare with the preset interval threshold in the real-time dynamic data acquisition device of the sensor;

[0007] S4: If the sample variance If the value is greater than or less than the preset threshold range, it is considered that the seat belt is not worn correctly, and the program will trigger an alarm.

[0008] The method for constructing the interval threshold includes the following steps S5: recruiting n volunteers; where n is a positive integer; each volunteer wears a safety belt;

[0009] S6: Each volunteer completes the following actions within the predetermined time: weightlifting, bending over, squatting, walking, and normal work. After completing the above actions, the data in the real-time dynamic sensor data acquisition device is as follows: ;

[0010] S7: Standardize the discrete data, where N represents the number of observations for each sample;

[0011] S8: The method for processing discrete data is: data extraction is performed on a per-observation basis, that is, in ··· In the process Sampling will produce N sets of data about... Data of length N-1 is defined as follows:

[0012] Where j represents the number of samples. [1,N];

[0013] Perform N sets of data Processing yields the observed samples defined as follows: ;

[0014] According to the central limit theorem, ~N( , ),in( for ··· The overall mean, for ··· The overall variance;

[0015] Then, to Standardization process: , obtain new data ; Find it again sample mean Sample variance This means completing the construction of the interval threshold;

[0016] The sample variance The construction method is as follows: the computer chip inside the real-time dynamic data acquisition device of the sensor observes the acquired data and generates an observation set. ··· , ··· Normalize it The data is compressed to the (-2,2) interval and matched with the constructed interval threshold, and the dimensions are eliminated; thus, the sample variance of the real-time pressure observation data is obtained. ;Including step S9: The sensor data real-time dynamic acquisition device captures the initial pressure value of the waist of the seat belt, and the captured value is d;

[0017] When the sample variance When the value exceeds the interval threshold, a test is performed. = And continue to observe;

[0018] like , and If the result is d, the program will issue an alarm for not wearing a seatbelt.

[0019] As a preferred technical solution of the present invention, it includes step S10: if it is not d at the same time, then proceed to the process of dynamic calculation of the variance of the real-time pressure observation data sample.

[0020] As a preferred technical solution of the present invention, step S11: the acquisition method of d is: the sensor data real-time dynamic acquisition device is connected to collect the real-time data after the pressure sensor is turned on, and forms an initial value d;

[0021] As a preferred technical solution of the present invention, it includes step S12: if the sample variance When the upper and lower limits of the range threshold are exceeded, the capture method of d is as follows: starting from the acquisition of the outlier, the first outlier is taken as the starting point, and the first outlier and the next two pressure values ​​are taken as a test sample group for secondary test, and finally it is determined whether the seat belt has been removed.

[0022] As a preferred technical solution of the present invention, it includes step S13: using a formula during the secondary inspection. = Perform the calculation, if Then it is believed , and If the value is simultaneously checked as d, then the first outlier point will be identified as the point that has reached the alarm value.

[0023] As a preferred technical solution of the present invention, the range of the interval threshold is -0.05 to +0.06.

[0024] As a preferred technical solution of the present invention, it further includes the following step, S13: if the sample variance constructed by the real-time dynamic acquisition device of sensor data... When the value is less than or greater than the range threshold, an alarm is issued, and the monitoring signal is transmitted to the repeater via a signal transmitting device connected to the real-time dynamic data acquisition device of the sensor. The repeater then transmits the signal to the router set up in the construction site, and finally the router transmits it to the on-site safety department.

[0025] The beneficial effects of the present invention by adopting the above technical solution are as follows: Since the device sets an interval threshold inside the data collector, as long as the acquired data is within the interval threshold, it can be determined that the seat belt is worn correctly, reducing the probability and number of false alarms, and avoiding false alarms from startling users or causing user resentment. Attached Figure Description

[0026] Figure 1 A flowchart illustrating the workflow of a pressure sensor and data acquisition unit;

[0027] Figure 2 Dynamic pressure value plots are constructed for interval thresholds;

[0028] Figure 3 Normalized plot of dynamic pressure values ​​when constructing interval thresholds;

[0029] Figure 4 A variance and mean plot of dynamic pressure data constructed for interval thresholds;

[0030] Figure 5 A probability density curve of dynamic pressure data is constructed for the interval threshold.

[0031] Figure 6 This is a diagram showing the dynamic pressure values ​​during the on-site simulation.

[0032] Figure 7 This is a normalized graph of dynamic pressure values ​​during on-site simulation.

[0033] Figure 8 A plot showing the variance and mean of dynamic pressure data during on-site simulation;

[0034] Figure 9 This is a probability density curve of dynamic pressure data during on-site simulation. Detailed Implementation

[0035] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0036] Please see Figure 1-9 , specifically Figure 1 As shown, one embodiment of the present invention provides a method for detecting the correct wearing of a seat belt, including a data acquisition device installed on the waist of the seat belt, and a preset interval threshold in the data acquisition device; wherein, the data acquisition of the data acquisition device includes the following steps: S1: continuously acquiring pressure values ​​after being powered on by a pressure sensor installed on the waist of the seat belt, and pre-storing the pressure values ​​into a real-time dynamic acquisition device for sensor data in the data acquisition device;

[0037] S2: The real-time dynamic sensor data acquisition unit compares the acquired pressure value with its internal preset interval threshold; wherein, the interval threshold range is -0.05 to +0.06.

[0038] S3: If the variance corresponding to the pressure value exceeds the interval threshold range, the program determines it to be an alarm. For example, the variance calculated by the real-time dynamic acquisition device of sensor data is compared with the variance threshold obtained from the training samples. If the alarm condition is met and the number of consecutive values ​​is greater than 3, an alarm will be triggered; that is, more than 3 data points are considered to be a continuous triggering state, and an alarm will be triggered.

[0039] During construction work, workers will wear safety harnesses to perform a series of actions, resulting in various body postures such as bending over, lifting weights, bending to the side, and squatting. Because the movement of workers' bodies is dynamic, continuous, and random, setting a fixed threshold in the data acquisition device to judge and alarm based on sensor pressure values ​​will lead to numerous false alarms. For example, during weightlifting, the pressure sensor located at the waist will receive a reduced pressure value due to the arched posture; using a fixed threshold for judgment would result in false alarms.

[0040] Another point is that the pressure sensor is a portable thin-film type that measures pressure through resistance. Its initial value after each power-on will change with the stability of the environment, humidity, and other indicators, making it impractical to use a fixed threshold. If the difference between the measured data and the initial value is used as the threshold, the threshold may change due to the initial value being too high or too low.

[0041] To eliminate the influence of various movement postures of wearers on the alarm determination, the inventors categorized the various movement postures of different body types into a "threshold shift of individual behavior on sample distribution morphology." By establishing a method for observing the impact of individual behavior on the threshold shift of sample distribution morphology, we first constructed a training set.

[0042] When constructing interval thresholds

[0043] Because the operator's physique, height, weight, and operating habits can all affect the sensor's data interpretation, the inventor recruited n individuals, where n is a positive integer; the more n individuals, the better.

[0044] For example, a training set can be constructed by collecting data from 100 test volunteers while each person wears a seatbelt for 40 minutes. During the data collection period, the test participants are required to perform basic actions such as weightlifting, bending over, squatting, walking, and normal work.

[0045] We will express the test volunteer data and methods in statistical terms as follows: For observation data with n observers... Since the data is relatively discrete, it is standardized, where N represents the number of observations for each sample.

[0046] The processing method involves data extraction on a per-observation-individual basis, that is, in ··· In the process Sampling will produce N sets of data about... Data of length N-1 is defined as

[0047] Where j represents the number of samples. [1,N];

[0048] Perform N sets of data Processing yields the observed samples defined as follows: .

[0049] According to the central limit theorem, ~N( , ),in( for ··· The overall mean, for ··· The overall variance. Then, for Standardization process: , obtain new data .

[0050] After the above processing, the result can be calculated again. sample mean Sample variance This means completing the construction of the interval threshold.

[0051] To minimize false positives, the sample variance in the training set is adjusted. The upper and lower limits are used for training observations to obtain the maximum and minimum deviations between the pressure value observed by the pressure sensor and the average value observed during the process when the worker is wearing a seat belt.

[0052] The measured threshold range varies to some extent, as detailed below:

[0053] When the number of detected pressure values ​​is less than 20, the threshold range is "-0.2 to +0.3";

[0054] When the number of detected pressure values ​​is between 20 and 50, the threshold range is "-0.16 to +0.18";

[0055] When the number of detected pressure values ​​is between 51 and 100, the threshold range is "-0.1 to +0.12";

[0056] When the number of detected pressure values ​​is between 101 and 200, the threshold range is "-0.08 to +0.09";

[0057] When the number of pressure values ​​being detected is above 200, the threshold range is "-0.05 to +0.06".

[0058] The above observations show that when the sample size exceeds 200, the threshold changes relatively little as the sample size increases. A threshold range of -0.05 to +0.06 satisfies the discrimination requirement, which is an accuracy of over 90%. Therefore, the more values ​​of n, the better.

[0059] During the on-site simulation

[0060] After the sample is constructed, during actual construction operations, personnel wearing pressure sensor equipment will return to the site to observe the observation set in real time.

[0061] Real-time on-site observations will generate observation sets. ··· , ··· Normalize it. Since this transformation does not change the distribution points of the pressure observation curve, it only serves to compress the data to the (-2,2) interval, match it with the data interval of the training set, and eliminate the dimensions.

[0062] The real-time dynamic data acquisition device calculates the sample variance of the pressure values ​​observed by the pressure sensor in real time. When the sample variance When the activity exceeds the upper or lower limit of the interval threshold, it is considered that the worker wearing the safety belt is acting abnormally, and thus it is necessary to check whether the worker has removed the safety belt.

[0063] when If the limit is exceeded, an inspection shall be conducted. = And continue to observe, if , and If both are checked and found to be d, the program will issue a seatbelt not being worn alarm.

[0064] If they are not both d, then re-enter the sample variance of the pressure values ​​observed by the pressure sensor. The dynamic calculation process minimizes false alarms and utilizes deep learning to investigate the impact of various behaviors of seatbelt-wearing workers on pressure observations. The reasons for deviations from N(0,1) are, for example, as the number of users continues to increase, the interval threshold is calculated in real time in the above manner, so that the device can correct the interval threshold in real time to reduce the number of false alarms of seat belts.

[0065] It should be noted that, in order to further optimize the algorithm and further reduce the false alarm rate of the device, this invention optimizes the algorithm in the following way:

[0066] The sensor signal response status is analyzed in two stages to determine whether an alarm is triggered.

[0067] 1. Preliminary assessment stage

[0068] 1) Initial value capture

[0069] Due to the influence of humidity, temperature, etc., the initial pressure value will change. Therefore, the program is set to capture the initial pressure value a short period of time before wearing the seat belt. The captured value is d.

[0070] For example, after powering on, data is captured every second for 3 seconds, with the lowest value used as the initial value. Due to the influence of humidity, temperature, etc., the initial pressure value may change. Therefore, the program is set to capture the initial pressure value for a short period before wearing the seat belt, with the capture value being 'd'. That is, after the sensor is powered on or after an alarm is triggered, and the person puts on the seat belt again or tightens it until the alarm is not triggered, 3 pressure values ​​are captured within the first 3 seconds, and the minimum value is used as the initial value. For example, after powering on, data is captured every second for 3 seconds, with the lowest value used as the initial value.

[0071] 2) Calculate the variance of real-time data

[0072] Real-time on-site observations will generate observation sets. ··· , ··· The seatbelt lumbar pressure values ​​captured in real time by the sensor are normalized. Since this transformation does not change the distribution points of the pressure observation curve, it only serves to compress the data to the (-2,2) interval, match it with the data interval of the training set, and eliminate the dimensions.

[0073] The real-time dynamic data acquisition device calculates the sample variance of the pressure values ​​observed by the pressure sensor in real time. ,when When the activity exceeds the upper or lower limit of the interval threshold, it is preliminarily determined that the worker wearing the safety belt is exhibiting abnormal activity.

[0074] Entering the second stage: Starting from the acquisition of outliers, the first outlier is used as the starting point. The first outlier and the next two pressure values ​​are used as a test sample group for secondary testing to finally determine whether the seat belt has been removed.

[0075] 2. Final Judgment Stage

[0076] The point where the variance of the obtained pressure value exceeds the upper limit of the variance threshold is taken as the first outlier. The first outlier, along with the second and third pressure values, form a sample group for a secondary test. The formula is used... = Perform the calculation, if Then it is believed , and If the value is simultaneously checked and found to be d, then the first outlier point will be identified as the point that has reached the alarm value. In the formula, d is the initial pressure value initially captured.

[0077] In summary, the entire discrimination method actually employs two approaches: the first stage uses the variance calculated from real-time pressure data for preliminary verification; the second stage uses the difference between the measured pressure value and the initial value, i.e., the relative pressure value, for discrimination. The evaluation objects used in the two stages are different: the first stage uses the variance calculated from the pressure value, while the second stage uses the difference between the measured pressure value and the initial pressure value, i.e., the relative pressure value.

[0078] Based on this, the above method addresses two key issues: First, it solves the problem of initial pressure value variation when using variance for initial discrimination, i.e., it eliminates the influence of the initial pressure value on real-time pressure data. For example, under certain temperature and humidity conditions, the initial pressure value may be higher than the lower value of later test data, meaning the initial pressure value is not the minimum pressure value throughout the entire test. Second, it solves the problem that directly using the difference between the measured pressure value and the initial value (i.e., the relative pressure value) for discrimination can lead to errors in discrimination and an increased false alarm rate because the initial pressure value may change with environmental factors during continuous observation.

[0079] It is particularly important to note that when using the difference between the measured pressure value and the initial value (i.e., the relative pressure value) in the second stage, changes in the initial pressure value will not have any impact, because the discrimination object only includes three pressure value points, not all pressure value points. If the difference between the measured pressure value and the initial value (i.e., the relative pressure value) were used directly, it would apply to all dynamic measured data points, which could potentially cause unnecessary errors.

[0080] Algorithm testing and execution

[0081] After the test personnel put on safety harnesses, they simulated the working actions of real construction workers. The simulation data test showed that the sensor had good sensitivity to standing, bending over, unfastening the safety harness, and putting on the safety harness, with obvious changes in the image. Based on the actual construction situation on site, the test personnel were simulated twice, including lifting operations, bending over operations, and multi-person simulated construction operations.

[0082] Lifting Operation: The first simulated lifting operation included six actions: standing and walking, lifting operation, standing and walking again, lifting operation, loosening the safety belt, and fastening the safety belt. The simulated numerical test results showed a mean M = 2.0233 * 10^-5 and a variance σ = 0.0609. See details below. Figures 2-5 As shown.

[0083] Analysis of the data normalization graph reveals significant differences in sensor data values ​​under different postures. The characteristics of the data values ​​under different postures will be analyzed below.

[0084] Bending over: The second simulated bending over operation included six actions: standing and walking, bending over, standing and walking again, bending over, loosening the safety belt, and fastening the safety belt. The simulated numerical test results were a mean M = 0.0087 and a variance σ = 0.0626. Specific test data are as follows... Figures 6-9 As shown.

[0085] Simulated construction work with multiple personnel: The above tests show that the pressure sensor has good sensitivity to various actions of construction workers. However, to avoid false alarms caused by the sensor failing to detect pressure for a short time due to actions such as lifting or briefly adjusting safety belts during construction, further testing is needed.

[0086] Additionally, when lifting weights or bending over, the graph shows an oscillating waveform; when the safety belt is unfastened or not worn, the graph shows a straight line.

[0087] Based on the above plan, specifically as follows: Figure 1 As shown, the method also includes the following steps: if the pressure value is less than the interval threshold, an alarm is issued, and the monitoring signal is transmitted to the repeater through the signal transmitting device connected to the pressure sensor. The repeater then transmits the signal to the router set up in the construction site, and finally the router transmits the signal to the on-site safety department.

[0088] The above methods enable remote monitoring of seat belts, further reducing the occurrence of safety accidents caused by human negligence. In particular, the ability to issue alarm sounds helps users raise their safety awareness and allows inspectors to focus on monitoring users who violate regulations.

[0089] Additionally, the pressure sensor parameters are as follows: Outer material: Black; Encapsulation: Elastic cloth; Internal: Elastic sensitive material; Standard dimensions: Radius 30 mm, Sensitive area 20 mm, Thickness 2.3 mm radius; Surrounding area can be sewn or glued. Allowable pressure range: 0~150N; Resolution: Less than 0.05N; Durability: Over 300,000 cycles; Response time: <20ms; Operating temperature: <60℃.

[0090] The parameters of the real-time dynamic sensor data acquisition unit are as follows: Motion posture sensing module: ADXL345BCCZ; Purpose: Used for sensing motion state (vibration state and activity state detection);

[0091] Detailed parameters:

[0092] Ultra-low power consumption: 23μA in measurement mode and 0.1μA in standby mode at VS=2.5V (typical value). Power consumption automatically scales proportionally with bandwidth. Vibration and activity (active / inactive) detection, free fall detection. Power supply voltage range: 2.0V to 3.6V; I / O voltage range: 1.7V to VS; SPI bandwidth. Temperature range (-40°C to +85°C); Dimensions: 3mm×5mm×1mm, LGA package; Module size: 23*18*11mm; Capacitance digital converter: PCAP01AD. Application: Capacitance measurement; Detailed parameters: Operating temperature: -40°C to +125°C; Power supply voltage... 2.1V~3.6V; Package: QFN-32; Battery Fuel Gauge: MAX17048G+T10; Application: Power measurement; Detailed parameters: Operating temperature: -40℃ to +85℃; Power supply voltage: 2.4V~4.5V; Package: TDFN-EP; Chip interface: I2C; Electrically Erasable Programmable Read-Only Memory (EEPROM): K24C02; Application: Storage of configuration information and algorithms; Detailed parameters: Operating temperature: -40℃ to +85℃; Power supply voltage: 2.4V~4.5V; Package: TDFN-EP; Chip interface: I2C; EEPROM: K24C02; Application: Storage of configuration information and algorithms; Detailed parameters: Operating temperature: -40℃ to +85℃; Power supply voltage: 2.4V~4.5V; Voltage: 1.7V~5.5V; Package: TDFN-EP; Chip Interface: I2C; Lithium Battery Linear Charging Chip: FM4057E42; Application: Output overcurrent, overvoltage, and short circuit protection; Input voltage range: 4.6V~32V; Input overvoltage protection: 1) FM4057 overvoltage protection voltage is 6.5V; 2) FM4057B supports high-voltage charging; Supports OTP over-temperature protection with 4KV ESD characteristics; Package type: SOT23-6 package; External resistor adjusts charging current: Maximum support 1A charging; Supports 4.2V, 4.35V, and 4.4V battery types; Supports LED display of charging status, supports dual-color LED display; Built-in MOSFET with anti-reverse current function port; Main control chip: CC2530; CPU core model: 8051; Fast Erasable Programmable Read-Only Memory (FLASH): FLASH_SIZE: 256KB; Static Random Access Memory (SRAM): 8KB; Provides USB interface.

[0093] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. A method for detecting correct seatbelt wearing, characterized in that: Including a real-time dynamic sensor data acquisition device; The data acquisition of the real-time dynamic sensor data acquisition device includes the following steps: S1: After the pressure sensor located at the waist of the seat belt is powered on, it will continuously collect the pressure value of the seat belt and store the pressure value in the real-time dynamic sensor data acquisition device. S2: The real-time pressure data acquired by the sensor data acquisition device is used to construct a sample variance of the pressure values. ; S3: The real-time dynamic acquisition device for sensor data will construct the sample variance. Compare with the preset interval threshold in the real-time dynamic data acquisition device of the sensor; S4: If the sample variance If the value is greater than or less than the preset threshold range, it is considered that the seat belt is not worn correctly, and the program will trigger an alarm. The method for constructing the interval threshold, The steps included in S5 are as follows: Recruit n volunteers; where n is a positive integer; each volunteer wears a seatbelt; S6: Each volunteer completes the following actions within the predetermined time: weightlifting, bending over, squatting, walking, and normal work. After completing the above actions, the data in the real-time dynamic sensor data acquisition device is as follows: ; S7: Standardize the discrete data, where N represents the number of observations for each sample; S8: The method for processing discrete data is: data extraction is performed on a per-observation basis, that is, in ··· In the process Sampling will produce N sets of data about... Data of length N-1 is defined as follows: Where j represents the number of samples, [1,N]; Perform N sets of data Processing yields the observed samples defined as follows: ; According to the central limit theorem, ~N( , ),in for ··· The overall mean, for ··· The overall variance; Then, to Standardization process: , obtain new data ; Find it again sample mean Sample variance This means completing the construction of the interval threshold; The sample variance The construction method is as follows: the computer chip inside the real-time dynamic data acquisition device of the sensor observes the acquired data and generates an observation set. ··· , ··· Normalize it The data is compressed to the (-2,2) interval and matched with the constructed interval threshold, and the dimensions are eliminated; thus, the sample variance of the real-time pressure observation data is obtained. ; Step S9: The real-time dynamic sensor data acquisition device captures the initial pressure value of the waistband of the seat belt, and the captured value is d; When the sample variance When the value exceeds the interval threshold, a test is performed. = And continue to observe; like , and If the result is d, the program will issue an alarm for not wearing a seatbelt.

2. The method for detecting correct seatbelt wearing according to claim 1, characterized in that: Including step S10: If it is not d at the same time, then proceed to the real-time pressure observation data sample variance. The process of dynamic calculation.

3. The method for detecting correct seatbelt wearing according to claim 1, characterized in that: The method for capturing step S11:d is as follows: the sensor data real-time dynamic acquisition device is connected to collect the real-time data after the pressure sensor is turned on, and forms the initial value d.

4. The method for detecting correct seatbelt wearing according to claim 1, characterized in that: Including step S12: If the sample variance When the upper and lower limits of the interval threshold are exceeded, the capture method of d is as follows: starting from the acquisition of the outlier, the first outlier is taken as the starting point, and the first outlier and the two pressure values ​​thereafter are used as a test sample group for secondary testing, and finally it is determined whether the seat belt has been removed.

5. The method for detecting correct seatbelt wearing according to claim 1, characterized in that: Including step S13: Using the formula during the secondary test. = Perform the calculation, if Then it is believed , and If the value is simultaneously checked as d, then the first outlier will be identified as the point that has reached the alarm value.

6. The method for detecting correct seatbelt wearing according to claim 1, characterized in that: The threshold range is -0.05 to +0.

06.

7. The method for detecting correct seatbelt wearing according to claim 1, characterized in that: It also includes the following step, S13: If the sample variance constructed by the real-time dynamic acquisition device of sensor data... When the value is less than or greater than the range threshold, an alarm is issued, and the monitoring signal is transmitted to the repeater via a signal transmitting device connected to the real-time dynamic data acquisition device of the sensor. The repeater then transmits the signal to the router set up in the construction site, and finally the router transmits it to the on-site safety department.

Images