Operating personnel safety monitoring method, system and equipment based on intelligent safety helmet

Abstract

The invention discloses an operator safety monitoring method, system and device based on an intelligent safety helmet, and relates to the technical field of safety monitoring, and the method comprises the steps: carrying out the sensing monitoring through the intelligent safety helmet, and obtaining real-time sensing information; comparing and judging real-time physiological data in the real-time sensing information with a physiological data reference to obtain a real-time judgment result; when the real-time judgment result shows that the physiological state is normal, analyzing the real-time sensing information to obtain a real-time motion state of the operator and whether the real-time motion state is a preset state, and if yes, extracting real-time positioning coordinates in the real-time sensing information; if the intelligent safety helmet is not located in the preset safety area, a first early warning signal is sent out, and an audible and visual alarm in the intelligent safety helmet is activated. According to the invention, the technical problem that comprehensive real-time monitoring and regional safety early warning of the physiological state and the operation behavior of the operator cannot be realized in the prior art is solved, and the technical effect of improving the safety management and control efficiency and the emergency risk response capability of the operator on the operation site is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety monitoring, in particular to a worker safety monitoring method, system and device based on an intelligent safety helmet. BACKGROUND

[0002] In an industrial work environment, workers often face complex dangerous factors such as high temperature, high altitude, and harmful gas, and the work intensity is large and the environment is variable, which can easily lead to physiological abnormalities or misentry into dangerous areas. However, the traditional safety management method mainly relies on manual patrol or single sensor monitoring, and it is difficult to timely grasp the physiological state and actual position distribution of the workers, and lacks dynamic analysis and risk prediction ability for the behavior state of the workers. At the same time, there is a lack of effective means for intelligent identification and real-time warning of risk behaviors such as exceeding the safety area, long-time stillness or abnormal motion, which poses a safety hazard. SUMMARY

[0003] The present application provides a worker safety monitoring method, system and device based on an intelligent safety helmet, which is used to solve the technical problem that the physiological state and work behavior of workers cannot be comprehensively and real-time monitored and region safety warned in the prior art.

[0004] In view of the above problems, the present application provides a worker safety monitoring method, system and device based on an intelligent safety helmet.

[0005] In a first aspect of the present application, a worker safety monitoring method based on an intelligent safety helmet is provided, the method comprising:

[0006] The integrated perception component loaded in the intelligent safety helmet is used to perceive and monitor the worker to obtain real-time perception information; the real-time physiological data in the real-time perception information is compared and judged with the physiological data benchmark in the local processor to obtain a real-time judgment result, wherein the local processor is loaded in the intelligent safety helmet; when the real-time judgment result shows that the physiological state is normal, a motion state classifier is activated to analyze the real-time perception information to obtain the real-time motion state of the worker; it is judged whether the real-time motion state is a predetermined state, if yes, the real-time positioning coordinates in the real-time perception information are extracted; if the real-time positioning coordinates are not in a predetermined safety area, a first warning signal is issued, and a sound-light alarm in the intelligent safety helmet is activated based on the first warning signal.

[0007] In a second aspect of the present application, a worker safety monitoring system based on an intelligent safety helmet is provided, the system comprising:

[0008] The perception monitoring module is used for monitoring the worker by an integrated perception component loaded in the intelligent safety helmet to obtain real-time perception information; the contrast discrimination module is used for comparing and discriminating real-time physiological data in the real-time perception information with physiological data reference in a local processor to obtain a real-time discrimination result, wherein the local processor is loaded in the intelligent safety helmet; the analysis module is used for activating a motion state classifier to analyze the real-time perception information when the real-time discrimination result shows that the physiological state is normal to obtain real-time motion state of the worker; the coordinate extraction module is used for judging whether the real-time motion state is a predetermined state, and if yes, extracting real-time positioning coordinates in the real-time perception information; the early warning module is used for issuing a first early warning signal if the real-time positioning coordinates are not in a predetermined safety area, and activating a sound-light alarm in the intelligent safety helmet based on the first early warning signal.

[0009] In a third aspect, the present application provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the method for monitoring safety of workers based on an intelligent safety helmet.

[0010] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0011] The present application monitors a worker by an integrated perception component loaded in an intelligent safety helmet to obtain real-time perception information; compares and discriminates real-time physiological data in the real-time perception information with physiological data reference in a local processor to obtain a real-time discrimination result, wherein the local processor is loaded in the intelligent safety helmet; activates a motion state classifier to analyze the real-time perception information when the real-time discrimination result shows that the physiological state is normal to obtain real-time motion state of the worker; judges whether the real-time motion state is a predetermined state, and if yes, extracts real-time positioning coordinates in the real-time perception information; issues a first early warning signal if the real-time positioning coordinates are not in a predetermined safety area, and activates a sound-light alarm in the intelligent safety helmet based on the first early warning signal. The present application solves the technical problem that the physiological state and work behavior of a worker cannot be comprehensively monitored in real time and regional safety early warning in the prior art, intelligently discriminates and warns the state and position of a worker by integrating physiological monitoring, motion recognition and high-precision positioning in an intelligent safety helmet, and achieves the technical effects of improving safety management and control efficiency and sudden risk response capability of workers in a work site. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description only constitute some of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present application.

[0013] Figure 1 A flowchart of a safety monitoring method for workers based on an intelligent safety helmet is provided for the embodiments of the present application.

[0014] Figure 2 A structural schematic diagram of a safety monitoring system for workers based on an intelligent safety helmet is provided for the embodiments of the present application.

[0015] Figure 3 A structural schematic diagram of an exemplary electronic device is provided for the embodiments of the present application.

[0016] Legend of reference signs: bus 300, receiver 301, processor 302, transmitter 303, memory 304, bus interface 305, perception monitoring module 11, comparison and discrimination module 12, analysis module 13, coordinate extraction module 14, and early warning module 15. DETAILED DESCRIPTION

[0017] The present application provides a safety monitoring method, system and device for workers based on an intelligent safety helmet, aiming to solve the technical problem that the physiological state and working behavior of workers cannot be comprehensively and real-time monitored and regionally safety warned in the prior art. By integrating physiological monitoring, motion recognition and high-precision positioning on the intelligent safety helmet, the state and position of workers are intelligently discriminated and warned, so as to improve the technical effect of safety control efficiency and sudden risk response capability of workers in the working site.

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present application.

[0019] It should be noted that any variation of the terms "comprise" and "have" is intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to the process, method, product or device.

[0020] Embodiment one, as Figure 1As shown, the application provides a smart safety helmet-based worker safety monitoring method, which comprises:

[0021] Step S100: sensing monitoring of the worker by an integrated sensing component loaded in the smart safety helmet to obtain real-time sensing information.

[0022] Further, the method provided by the application embodiment further comprises:

[0023] The integrated sensing component comprises a physiological sensor, a motion sensor, and a dual-mode locator; wherein the physiological sensor at least comprises a heart rate monitor, a blood pressure monitor, and a temperature monitor, the motion sensor at least comprises a step frequency monitor, a step length monitor, and a speed monitor, and the dual-mode locator comprises a Beidou satellite locator and an ultra-wideband short-range locator.

[0024] In the application embodiment, the worker is sensed and monitored by the integrated sensing component loaded in the smart safety helmet. The integrated sensing component comprises a physiological sensor, a motion sensor, and a dual-mode locator.

[0025] Specifically, the physiological sensor comprises a heart rate monitor, a blood pressure monitor, and a temperature monitor. The heart rate monitor uses the photoplethysmography technology to detect the optical signal change of the blood flow under the skin and outputs the heart rate value in real time. The blood pressure monitor records the arterial pressure fluctuation by the oscillometric method to directly obtain the systolic pressure and diastolic pressure. The temperature monitor measures the skin temperature of the head in real time based on the thermistor principle.

[0026] The motion sensor comprises a step frequency monitor, a step length monitor, and a speed monitor. The step frequency monitor detects the periodic fluctuation of acceleration in the vertical direction by a three-axis acceleration sensor to output the number of steps per unit time. The step length monitor provides the actual step distance of each step according to the dynamic relationship between the step frequency and the acceleration amplitude. The speed monitor reads the product of the step frequency and the step length to reflect the moving speed of the worker in real time and judge whether the behavior state is within the normal operation range.

[0027] The dual-mode locator is composed of a Beidou satellite locator and an ultra-wideband short-range locator, which is used to continuously obtain the position coordinates of the worker. The Beidou satellite locator receives multiple navigation satellite signals and uses the pseudo-range solution method to calculate the position of the worker in the geographic coordinate system in real time; the ultra-wideband short-range locator communicates with multiple UWB base stations and uses the time difference of arrival algorithm to achieve centimeter-level positioning accuracy in a closed or complex structure environment.

[0028] The data collected by all sensors is integrated and formatted inside the intelligent safety helmet through a unified data processing procedure to form multi-dimensional, time-synchronized real-time sensing information. The real-time sensing information includes real-time physiological data, real-time motion data, and real-time positioning data. The real-time physiological data includes heart rate value, systolic pressure, diastolic pressure, and body surface temperature; the real-time motion data includes step frequency, step length, and speed; and the real-time positioning data includes geographic coordinates.

[0029] Further, the method provided in the application embodiment further includes, before comparing the real-time physiological data in the real-time sensing information with the physiological data benchmark in the local processor to obtain a real-time discrimination result:

[0030] collecting historical physiological data records of the worker; traversing the historical physiological data records based on a first physiological index to obtain a first historical index parameter time sequence, wherein the first physiological index is any one of predetermined physiological indexes, and the predetermined physiological indexes include heart rate, blood pressure, and temperature; analyzing the first historical index parameter time sequence to determine a first reasonable parameter range; and establishing the physiological data benchmark based on the first reasonable parameter range and pre-storing the physiological data benchmark to the local processor.

[0031] In the application embodiment, first, historical physiological data records of a target worker are obtained from a preset historical database. The historical database is a standard data storage structure and is stored according to worker identity identification. The data is derived from daily monitoring data continuously uploaded by the intelligent safety helmet, and the content includes original values such as heart rate, blood pressure (systolic pressure and diastolic pressure), and body surface temperature collected at each moment, with complete time stamp and personnel number.

[0032] Next, the historical physiological data records are traversed based on a first physiological index to obtain a first historical index parameter time sequence. Specifically, any one index is selected from predetermined physiological indexes as the first physiological index, for example, heart rate is selected as the analysis object. After selection, the heart rate values in all historical records are traversed in time sequence, and the results are arranged according to time to form the first historical index parameter time sequence. The first physiological index is any one of the predetermined physiological indexes, and the predetermined physiological indexes include heart rate, blood pressure, and temperature.

[0033] Then, the first historical index parameter time sequence is analyzed to determine a first reasonable parameter range. The first historical index parameter time sequence is processed by using an extreme value screening method, and the maximum value and the minimum value are extracted to define the upper and lower limits of the physiological index of the worker in the normal state.

[0034] After the parameter range is obtained, it is used to establish a physiological data benchmark based on the first reasonable parameter range. In addition to the first physiological indicator, the remaining predetermined physiological indicators (blood pressure and temperature) are analyzed by the same process, and finally a physiological data benchmark containing the reasonable range of all physiological indicators is constructed.

[0035] Finally, the constructed physiological data benchmark is pre-stored in the local processor.

[0036] Step S200: Compare the real-time physiological data in the real-time sensing information with the physiological data benchmark in the local processor to obtain a real-time judgment result, wherein the local processor is loaded in the intelligent safety helmet.

[0037] In the embodiment of the present application, the real-time physiological data in the real-time sensing information is compared and judged with the physiological data benchmark in the local processor to determine whether the physiological state of the current worker is within the normal range. The physiological data benchmark is an individualized physiological indicator reference range constructed for the corresponding worker and pre-stored in the local processor inside the intelligent safety helmet. The processor serves as the core computing unit of the intelligent safety helmet and has the ability of real-time reading and logical judgment.

[0038] When performing the judgment, each physiological indicator value (such as heart rate, blood pressure, and temperature) in the real-time physiological data is compared item by item to determine whether it falls within the range set by the corresponding physiological data benchmark. If all indicators are within the range, the real-time judgment result is normal. If any indicator exceeds the upper or lower limit of the physiological data benchmark, the judgment result is an abnormal state.

[0039] Further, the method provided by the application embodiment further comprises:

[0040] After comparing the real-time physiological data in the real-time sensing information with the physiological data benchmark in the local processor to obtain a real-time judgment result, when the real-time judgment result shows that the physiological state is abnormal, a second warning signal is issued, and the audible and visual alarm in the intelligent safety helmet is activated based on the second warning signal.

[0041] In the embodiment of the present application, after comparing the real-time physiological data in the real-time sensing information with the physiological data benchmark in the local processor to obtain a real-time judgment result, if the judgment result shows that the worker has an abnormal physiological state, a warning response process is immediately executed to ensure personnel safety and achieve rapid intervention.

[0042] Specifically, when the real-time discrimination result indicates that any physiological index such as heart rate, blood pressure or body temperature exceeds the reasonable parameter range defined by the physiological data baseline of the worker, the local processor automatically judges that the current state is abnormal. At this time, a second warning signal is generated and output, which is a pre-defined event trigger instruction indicating that the monitored physiological data has exceeded the safety threshold and immediate on-site prompting measures need to be taken.

[0043] The second warning signal is sent directly to the audible and visual alarm in the intelligent safety helmet through the local processor. The audible and visual alarm is a pre-warning device integrated in the safety helmet, including a buzzer and a high-brightness LED indicator light, which can simultaneously issue sound and flashing light prompts after receiving the activation signal. The alarm is controlled by an electrical signal driving mode to ensure that the worker himself and the surrounding personnel can perceive the abnormal state at the first time, facilitating further measures or triggering manual intervention.

[0044] Through this process, localized recognition, immediate warning and on-site response of physiological abnormal state are realized, ensuring that safety prompts can be triggered without relying on external communication when a person suddenly feels unwell, thereby improving the intelligent safety protection capability of the work site.

[0045] Step S300: When the real-time discrimination result shows that the physiological state is normal, activate the motion state classifier to analyze the real-time sensing information to obtain the real-time motion state of the worker.

[0046] In the embodiments of the present application, when the real-time discrimination result shows that the physiological state of the worker is normal, the motion behavior recognition process is entered. This process first extracts the real-time motion data in the real-time sensing information, and performs vectorization processing on the real-time motion data together with the real-time physiological data to form a unified real-time motion vector. Then, the activated motion state classifier is used to classify and analyze the vector to calculate the real-time motion index representing the behavior characteristics. Finally, by matching the index with the corresponding relationship of the pre-set classification standard, the current real-time motion state is output, such as non-motion state, first-level motion state or second-level motion state, etc.

[0047] Further, in the method provided by the embodiments of the present application, when the real-time discrimination result shows that the physiological state is normal, the motion state classifier is activated to analyze the real-time sensing information to obtain the real-time motion state of the worker, and the method further comprises:

[0048] extracting real-time motion data in the real-time sensing information; performing vectorization processing on the real-time motion data and the real-time physiological data to obtain a real-time motion vector; performing classification analysis on the real-time motion vector through the motion state classifier to obtain a real-time motion index; matching the motion state corresponding to the real-time motion index, denoted as the real-time motion state.

[0049] In the embodiments of the present application, when identifying the real-time motion state of the work personnel, first, real-time motion data in the real-time sensing information is extracted. The real-time motion data includes basic motion characteristics such as step frequency, step length and speed, which are collected by the step frequency monitor, the step length monitor and the speed monitor built in the intelligent safety helmet. By directly reading the original output value of each sensor at the current timestamp, and arranging it in a unified time sequence format, it is used as the basic data set for this behavior identification.

[0050] Subsequently, the real-time motion data and the real-time physiological data are vectorized. Specifically, first, each item of original data is subjected to numerical normalization processing, so that different dimension indicators (such as heart rate in bpm, speed in m / s) are uniformly scaled to the [0, 1] interval. The normalization method adopts the minimum-maximum normalization, that is, each data is linearly mapped according to its upper and lower limits in the preset physiological data reference and motion parameter range. Taking heart rate as an example, if the heart rate range set by the physiological data reference is 60-100 bpm, then the normalized value corresponding to the current heart rate value 80 bpm is (80-60) / (100-60)=0.5. After normalization, the normalized data items are arranged in a fixed order, including heart rate, blood pressure (systolic and diastolic pressure), body temperature, step frequency, step length and speed, etc., to form a seven-dimensional numerical vector (heart rate 1 dimension, blood pressure 2 dimensions, body temperature 1 dimension, step frequency 1 dimension, step length 1 dimension, speed 1 dimension). The structured vector is the real-time motion vector.

[0051] Then, the real-time motion vector is classified and analyzed by the motion state classifier. In this process, first, the motion vector database embedded in the motion state classifier is called, which contains a plurality of labeled standard motion vector samples, each sample corresponding to a specific motion state category and motion index. Then, the real-time motion vector is compared with all the standard vectors in the motion vector database one by one, and the Euclidean distance is calculated as the similarity measure. The smaller the distance value, the higher the similarity. The standard vector with the smallest distance is selected as the first motion vector, and the distance value corresponding to it is taken as the first motion similarity. If the similarity value reaches the set predetermined similarity limit value (i.e. the distance value is less than the preset threshold), the first motion index corresponding to the first motion vector is directly assigned as the real-time motion index of this time.

[0052] Finally, according to the position of the real-time motion index in the preset grading standard, the motion state corresponding to the real-time motion index is matched, and the real-time motion state is recorded. The motion state is usually divided into non-motion state, first-level motion state and second-level motion state. Among them, the non-motion state means that the worker is in a static state, and there is no obvious walking or moving behavior, such as standing, sitting, waiting or long time inaction. The first-level motion state means that the worker is in a low-intensity activity state, such as slow walking, inspection, routine operation behavior, etc. The second-level motion state means that the worker is in a high-intensity motion state, such as rapid movement, running, emergency evacuation, and other abnormal behaviors or high-load operation conditions.

[0053] Further, the method provided by the application embodiment further comprises:

[0054] obtaining a motion vector database embedded in the motion state classifier; comparing the real-time motion vector with a first motion vector in the motion vector database to obtain a first motion similarity; if the first motion similarity reaches a predetermined similarity limit value, taking a first motion index corresponding to the first motion vector as the real-time motion index.

[0055] In the application embodiment, first, a motion vector database embedded in the motion state classifier is obtained. The motion vector database is a pre-constructed behavior sample set, which contains a plurality of vector samples representing standard behavior states. Each vector sample is composed of normalized multi-dimensional data, and the corresponding motion state level and its numerical expression, i.e., motion index, have been manually labeled.

[0056] Then, the real-time motion vector is compared with a first motion vector in the motion vector database. The comparison method is similarity calculation, and the Euclidean distance is used as the similarity judgment criterion. The real-time motion vector is taken as the input vector, and the Euclidean distance value is calculated between the input vector and each known vector in the motion vector database. By traversing all samples, the standard sample with the smallest distance from the current input vector is identified and recorded as the first motion vector. The distance value between the two is converted into a similarity value, which is defined as the first motion similarity. Similarity value = 1 / (1+distance value).

[0057] Subsequently, it is judged whether the first motion similarity reaches a predetermined similarity limit value. The predetermined similarity limit value is a threshold value set in the classifier, which is used to ensure the reliability of the matching result and is pre-set by a technical expert. For example, when the similarity value (i.e., distance) is less than 0.2, it means that the matching is valid. If the similarity meets the threshold condition, it means that the current behavior is highly similar to the state represented by the first motion vector. At this time, the first motion index corresponding to the first motion vector is taken as the real-time motion index.

[0058] Step S400: judging whether the real-time motion state is a predetermined state, and if so, extracting a real-time positioning coordinate in the real-time sensing information.

[0059] In the embodiment of the present application, after the identification of the real-time motion state of the worker is completed, it is judged whether the real-time motion state is a predetermined state. The predetermined state refers to a non-motion state, i.e., the worker is in a stationary state without obvious walking or moving behavior. If the real-time motion state is determined to be a non-motion state, a real-time positioning coordinate in the real-time sensing information is extracted. The positioning coordinate is provided by the data synchronously collected in the real-time sensing information, and specifically includes spatial coordinate information obtained by a Beidou satellite positioning meter and an ultra-wideband short-range positioning meter.

[0060] Step S500: if the real-time positioning coordinate is not in a predetermined safe area, a first warning signal is issued, and a sound-light alarm in the intelligent safety helmet is activated based on the first warning signal.

[0061] Further, the method provided by the embodiment of the application further comprises:

[0062] The predetermined safe area includes an area in the work area except for fixed dangerous sources and mobile dangerous sources.

[0063] In the embodiment of the present application, the extracted real-time positioning coordinate is first compared in space. If the real-time positioning coordinate is not in the predetermined safe area, i.e., the coordinate position is not within the set safe work range, the warning response process will be started immediately.

[0064] The predetermined safe area is a compliance work range pre-set based on the on-site work environment, which explicitly excludes areas with safety hazards. The area is delimited by a management personnel according to the spatial layout, and is usually input in the form of an electronic map, including boundary coordinate points or polygon area data. Among them, the influence range of fixed dangerous sources (such as high-voltage equipment, rotating machinery, open containers, etc.) and mobile dangerous sources (such as forklifts, hoisting vehicles, track transportation devices, etc.) is explicitly excluded, to ensure that only the effective space for the worker to safely stay and move is reserved. For example, in a metallurgical plant, the 5-meter radius range around the high-temperature furnace body and the transportation vehicle running channel are both delimited as non-enterable areas and are not included in the predetermined safe area.

[0065] The obtained real-time positioning coordinate is compared with the boundary coordinate of the predetermined safe area, and a space inclusion algorithm is used for judgment, i.e., whether the current coordinate point is included in the safe polygon area. If the judgment result is no, i.e., the coordinate point is located outside the dangerous area, the forbidden area or the delimited temporary construction range, a first warning signal will be issued immediately. The warning signal is a pre-defined alarm triggering instruction, which is used to notify the worker or the background of the illegal stay or abnormal approaching behavior.

[0066] Then, based on the first early warning signal, the audible and visual alarm in the intelligent safety helmet is activated. The audible and visual alarm is integrated inside the intelligent safety helmet, including a buzzer device and a high-brightness LED lamp, which emits high-frequency sound and flashing warning light by receiving the control signal, realizes the instant reminding of the worker himself, and warns the surrounding workers to pay attention to the safety risk on site.

[0067] Further, the method provided by the application embodiment further comprises:

[0068] If the real-time motion state is not the predetermined state, the image collector in the intelligent safety helmet is activated; the motion image time sequence of the worker is acquired through the image collector; the motion trajectory obtained by analyzing the motion image time sequence is traversed in the historical motion trajectory record of the worker to obtain the most similar historical trajectory; the most similar historical trajectory is taken as the real-time predicted trajectory of the worker; if the real-time predicted trajectory is not in the predetermined safety area, a third early warning signal is issued, and based on the third early warning signal, the audible and visual alarm in the intelligent safety helmet is activated.

[0069] In the application embodiment, when the real-time motion state of the worker is determined to be a non-predetermined state, i.e., the current state is a first motion state or a second motion state, it means that the worker is in continuous movement or high-intensity exercise, at this time, in order to realize the predictive monitoring of the behavior trend, the visual perception means is enabled for dynamic tracking and risk judgment.

[0070] Specifically, first, the image collector in the intelligent safety helmet is activated. The image collector is a small high-definition camera integrated in the front or side of the intelligent safety helmet, which has automatic focusing and wide-angle shooting capability. After activation, the device continuously acquires the working environment images in front of the worker at a set frame rate (for example, 10 frames per second), generates a time-continuous image data sequence, and forms a motion image time sequence. The image is cached to the local processor in JPEG or YUV format.

[0071] Then, the motion image time sequence is dynamically identified by an image analysis method, the change path of the position of the worker in the continuous image frames is extracted, and the motion trajectory of the current period is constructed. This step uses a time sequence displacement extraction technology based on the optical flow method, specifically uses the Lucas-Kanade optical flow algorithm, tracks the displacement change of the feature points (such as ground markers, equipment edges, and other fixed references) in the image between frames, and calculates the displacement vector of the feature points between adjacent frames. By accumulating the displacement vectors of multiple feature points and combining the positioning information of the intelligent safety helmet, a set of relative motion vectors between frames is output, and an actual motion trajectory in two-dimensional space is generated by combination.

[0072] The extracted motion trajectory is then compared with a pre-stored historical motion trajectory record of the worker. The historical trajectory record is generated according to actual positioning data during previous work of the worker, is stored in a classified manner, and is coded according to trajectory segments. The comparison is performed through a traversal method based on trajectory similarity evaluation, using a Hausdorff distance calculation method, matching historical trajectory samples one by one, calculating the Hausdorff distance between the current trajectory and each historical trajectory, identifying the trajectory segment with the smallest distance as the most similar historical trajectory, and representing the current behavior trend as the most similar trajectory in the known behavior mode in the past.

[0073] At this time, the most similar historical trajectory is taken as a real-time prediction trajectory of the current worker, and is used to infer a potential area to which the worker will move.

[0074] Next, spatial position analysis is performed on the real-time prediction trajectory to determine whether the worker is not in a predetermined safe area. The predetermined safe area is an operation-permitted area set according to a work site plan, is described by polygon coordinate data, and excludes spaces occupied by fixed dangerous sources (such as high-voltage equipment and mechanical arm operation areas) and mobile dangerous sources (such as track transport vehicle operation paths). The determination method uses spatial overlap detection based on a point set trajectory and a polygon boundary to determine whether the trajectory has a crossing or extension behavior. If the determination result is that the real-time prediction trajectory is not in the predetermined safe area, that is, the worker has a trend of deviating from the safe area or entering a risk area, a third warning signal is immediately sent out.

[0075] Finally, the audible and visual alarm in the intelligent safety helmet is activated based on the third warning signal. After receiving the third warning signal, the alarm is started, the buzzer emits a continuous high-frequency sound, and the LED light continuously emits light in a flashing mode, thereby issuing a visual and audible warning to the worker and surrounding personnel, prompting the existence of a position risk and guiding the worker to leave the non-safe area in time.

[0076] In the embodiments of the present application, the above-mentioned embodiments of the present application have at least the following technical effects:

[0077] This application uses an integrated sensing component installed in a smart safety helmet to monitor and perceive workers, obtaining real-time sensing information. The real-time physiological data in the sensing information is compared and judged with a physiological data benchmark in a local processor to obtain a real-time judgment result. The local processor is installed in the smart safety helmet. When the real-time judgment result indicates a normal physiological state, a motion state classifier is activated to analyze the real-time sensing information to obtain the worker's real-time motion state. It is determined whether the real-time motion state is a predetermined state; if so, the real-time positioning coordinates in the sensing information are extracted. If the real-time positioning coordinates are not within a predetermined safe area, a first warning signal is issued, and an audible and visual alarm in the smart safety helmet is activated based on the first warning signal. This invention solves the technical problem in the prior art of being unable to achieve comprehensive real-time monitoring and area safety warning of workers' physiological state and work behavior. By integrating physiological monitoring, motion recognition, and high-precision positioning in a smart safety helmet, it intelligently judges and warns of workers' state and location, achieving the technical effect of improving the efficiency of on-site personnel safety management and the ability to respond to sudden risks.

[0078] Example 2 is based on the same inventive concept as the worker safety monitoring method based on smart safety helmets in the previous examples, such as... Figure 2 As shown, this application provides a worker safety monitoring system based on a smart safety helmet. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0079] The system comprises: a perception and monitoring module 11, used to perceive and monitor the worker through an integrated perception component installed in the smart safety helmet, and obtain real-time perception information; a comparison and discrimination module 12, used to compare and discriminate the real-time physiological data in the real-time perception information with the physiological data benchmark in the local processor, and obtain a real-time discrimination result, wherein the local processor is installed in the smart safety helmet; an analysis module 13, used to activate a motion state classifier to analyze the real-time perception information and obtain the real-time motion state of the worker when the real-time discrimination result shows that the physiological state is normal; a coordinate extraction module 14, used to determine whether the real-time motion state is a predetermined state, and if so, extract the real-time positioning coordinates from the real-time perception information; and an early warning module 15, used to issue a first early warning signal if the real-time positioning coordinates are not in a predetermined safe area, and to activate the audible and visual alarm in the smart safety helmet based on the first early warning signal.

[0080] Furthermore, the system is also used to implement the following functions:

[0081] The integrated perception component comprises a physiological sensor, a motion sensor and a dual-mode locator; wherein the physiological sensor comprises at least a heart rate monitor, a blood pressure monitor and a temperature monitor, the motion sensor comprises at least a step frequency monitor, a step length monitor and a speed monitor, and the dual-mode locator comprises a Beidou satellite locator and an ultra-wideband short-range locator.

[0082] Further, the system is also used to realize the following functions:

[0083] Collecting historical physiological data records of the worker; traversing the historical physiological data records based on a first physiological index to obtain a first historical index parameter time sequence, wherein the first physiological index is any one of predetermined physiological indexes, and the predetermined physiological indexes include heart rate, blood pressure and temperature; analyzing the first historical index parameter time sequence to determine a first reasonable parameter range; and establishing the physiological data benchmark based on the first reasonable parameter range and pre-storing the physiological data benchmark to the local processor.

[0084] Further, the system is also used to realize the following functions:

[0085] After comparing and judging the real-time physiological data in the real-time perception information with the physiological data benchmark in the local processor to obtain a real-time judgment result, when the real-time judgment result shows that the physiological state is abnormal, a second warning signal is sent, and the audible and visual alarm in the intelligent safety helmet is activated based on the second warning signal.

[0086] Further, the system is also used to realize the following functions:

[0087] Extracting real-time motion data in the real-time perception information; vectorizing the real-time motion data and the real-time physiological data to obtain a real-time motion vector; classifying and analyzing the real-time motion vector through the motion state classifier to obtain a real-time motion index; and matching the motion state corresponding to the real-time motion index, which is recorded as the real-time motion state.

[0088] Further, the system is also used to realize the following functions:

[0089] Obtaining a motion vector database embedded in the motion state classifier; comparing the real-time motion vector with a first motion vector in the motion vector database to obtain a first motion similarity; and if the first motion similarity reaches a predetermined similarity limit value, taking a first motion index corresponding to the first motion vector as the real-time motion index.

[0090] Further, the system is also used to realize the following functions:

[0091] determining whether the real-time motion state is a predetermined state, if not, activating an image collector in the intelligent safety helmet; acquiring a motion image time sequence of the worker through the image collector; traversing a motion trajectory obtained by analyzing the motion image time sequence in a historical motion trajectory record of the worker to obtain a most similar historical trajectory; taking the most similar historical trajectory as a real-time predicted trajectory of the worker; if the real-time predicted trajectory is not in the predetermined safety region, issuing a third warning signal, and activating an audible and visual alarm in the intelligent safety helmet based on the third warning signal.

[0092] Further, the system is also used to implement the following functions:

[0093] The predetermined safety region includes a region in the work area except for fixed and mobile dangerous sources.

[0094] Embodiment three, based on the inventive concept of the work personnel safety monitoring method based on the intelligent safety helmet in the foregoing embodiments, the present application also provides an electronic device, comprising: at least one processor; a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the steps of any one of the method in the foregoing embodiment one.

[0095] Figure 3 The structure schematic diagram of the exemplary electronic device of the present application is shown in FIG. 1. Figure 3 In FIG. 1, the bus architecture is represented by a bus 300, which can include any number of interconnecting buses and bridges, the bus 300 connecting various circuits including one or more processors represented by a processor 302 and memory represented by a memory 304. The bus 300 can also connect various other circuits such as peripheral devices, voltage stabilizers and power management circuits, which are well known in the art, and therefore, will not be further described herein. A bus interface 305 provides an interface between the bus 300 and a receiver 301 and a transmitter 303. The receiver 301 and the transmitter 303 can be the same element, i.e. a transceiver, which provides a unit for communicating with various other devices on a transmission medium. The processor 302 is responsible for managing the bus 300 and general processing, while the memory 304 can be used to store data used by the processor 302 in performing operations.

[0096] It should be noted that the above-mentioned embodiment sequence of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the above describes a specific embodiment of the present application. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0097] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0098] The present application is only an exemplary description of the present application, and is considered to cover any and all modifications, changes, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various modifications and changes to the present application without departing from the scope of the present application. Thus, if these modifications and changes of the present application belong to the scope of the present application and its equivalents, the present application intends to include these modifications and changes.

Claims

1. A method for monitoring safety of workers based on smart safety helmets, characterized in that, include: The integrated sensing components installed in the smart safety helmet are used to monitor workers and obtain real-time sensing information. The real-time physiological data in the real-time sensing information is compared and judged with the physiological data benchmark in the local processor to obtain the real-time judgment result, wherein the local processor is loaded in the smart safety helmet; When the real-time discrimination result shows that the physiological state is normal, the motion state classifier is activated to analyze the real-time perception information and obtain the real-time motion state of the worker. Determine whether the real-time motion state is a predetermined state; if so, extract the real-time positioning coordinates from the real-time sensing information. If the real-time positioning coordinates are not within the predetermined safe area, a first warning signal is issued, and the sound and light alarm in the smart safety helmet is activated based on the first warning signal. 2.The smart safety helmet based worker safety monitoring method according to claim 1, wherein, The integrated sensing component includes a physiological sensor, a motion sensor, and a dual-mode locator; The physiological sensors include at least a heart rate monitor, a blood pressure monitor, and a temperature monitor; the motion sensors include at least a cadence monitor, a step length monitor, and a speed monitor; and the dual-mode locator includes a BeiDou satellite locator and an ultra-wideband short-range locator. 3.The smart safety helmet based worker safety monitoring method of claim 2, wherein, Before comparing and judging the real-time physiological data in the real-time sensing information with the physiological data benchmark in the local processor to obtain the real-time judgment result, the process includes: Collect historical physiological data records of the workers; The historical physiological data records are traversed based on the first physiological indicator to obtain the time series of the first historical indicator parameters. The first physiological indicator refers to any one of the predetermined physiological indicators, and the predetermined physiological indicators include heart rate, blood pressure and temperature. Analyze the time series of the first historical indicator parameters to determine the first reasonable parameter range; The physiological data benchmark is constructed based on the first reasonable parameter range and pre-stored in the local processor. 4.The smart safety helmet based worker safety monitoring method of claim 1, wherein, After comparing and judging the real-time physiological data in the real-time sensing information with the physiological data benchmark in the local processor to obtain the real-time judgment result, when the real-time judgment result shows that the physiological state is abnormal, a second warning signal is issued, and the sound and light alarm in the smart safety helmet is activated based on the second warning signal. 5.The smart safety helmet based worker safety monitoring method of claim 2, wherein, When the real-time discrimination result shows that the physiological state is normal, the motion state classifier is activated to analyze the real-time perceived information to obtain the real-time motion state of the worker, including: Extract real-time motion data from the real-time sensing information; The real-time motion data and the real-time physiological data are vectorized to obtain real-time motion vectors; The real-time motion vector is classified and analyzed by the motion state classifier to obtain the real-time motion index; The motion state corresponding to the real-time motion index is recorded as the real-time motion state. 6.The smart safety helmet based worker safety monitoring method according to claim 5, wherein, The real-time motion vector is classified and analyzed using the motion state classifier to obtain a real-time motion index, including: Obtain the motion vector database embedded in the motion state classifier; The real-time motion vector is compared with the first motion vector in the motion vector database to obtain the first motion similarity. If the first motion similarity reaches a predetermined similarity limit value, a first motion index corresponding to the first motion vector is taken as the real-time motion index. 7.The smart safety helmet based worker safety monitoring method of claim 1, wherein, Further comprising: judging whether the real-time motion state is a predetermined state, if not, activating an image collector in the intelligent safety helmet; acquiring a motion image time sequence of the worker through the image collector; traversing a motion trajectory obtained by analyzing the motion image time sequence in a historical motion trajectory record of the worker to obtain a most similar historical trajectory; taking the most similar historical trajectory as a real-time predicted trajectory of the worker; if the real-time predicted trajectory is not in the predetermined safety region, issuing a third warning signal and activating an audible and visual alarm in the intelligent safety helmet based on the third warning signal. 8.The smart safety helmet based worker safety monitoring method of claim 7, wherein, The predetermined safety region includes a region in the work area excluding fixed and mobile dangerous sources.

9. A worker safety monitoring system based on a smart safety helmet, characterized by, The system is used to execute the worker safety monitoring method based on the intelligent safety helmet as claimed in any one of claims 1-8, and the system comprises: a perception monitoring module for monitoring the worker through an integrated perception component loaded in the intelligent safety helmet to obtain real-time perception information; a comparison and discrimination module for comparing and discriminating real-time physiological data in the real-time perception information with physiological data benchmarks in a local processor to obtain real-time discrimination results, wherein the local processor is loaded in the intelligent safety helmet; an analysis module for activating a motion state classifier to analyze the real-time perception information when the real-time discrimination results show that the physiological state is normal to obtain a real-time motion state of the worker; a coordinate extraction module for judging whether the real-time motion state is a predetermined state, if so, extracting real-time positioning coordinates in the real-time perception information; a warning module for issuing a first warning signal if the real-time positioning coordinates are not in a predetermined safety region, and activating an audible and visual alarm in the intelligent safety helmet based on the first warning signal.

10. An electronic device, comprising: The electronic device comprises: a memory for storing executable instructions; a processor for executing the executable instructions stored in the memory to implement the worker safety monitoring method based on the intelligent safety helmet as claimed in any one of claims 1-8.

Images