A smart construction site personnel behavior monitoring and early warning method, device and equipment

Abstract

The application provides a kind of wisdom construction personnel behavior monitoring early warning method, device and equipment, method includes: obtaining the complete terrain data of target slope in target wisdom construction;The target slope is divided into multiple sub-regions;Obtain the image video frame sequence of target personnel when moving in sub-region in sliding time window;Get the center of gravity trajectory of target personnel when moving and the slope of sub-region;Determine the initial behavior safety index of target personnel behavior action;Obtain the wind speed time history data in sub-region;Calculate wind-induced risk coefficient;Get the ground reflection intensity data collected by multispectral sensor and the environmental data in sub-region;Calculate the wet and slippery risk coefficient;According to initial behavior safety index, wind-induced risk coefficient and wet and slippery risk coefficient, obtain comprehensive risk index;According to comprehensive risk index, the behavior action of target personnel in sub-region is early warned.The scheme of the application can monitor and early warn the personnel behavior of pumped storage power station construction site.

Description

Technical Field

[0001] This invention relates to the field of computer information technology processing, and in particular to a method, device, and equipment for monitoring and early warning of personnel behavior at smart construction sites. Background Technology

[0002] Pumped storage power station construction sites typically have significant elevation differences, consisting of a pumping tank located in a low-lying area and a storage tank located at a higher elevation, connected by a ramp to form a construction access road. The slope of this ramp from the pumping tank in the lower area to the storage tank in the higher area varies significantly with the terrain, forming a continuous and differentiated sloping structure. Personnel working on the ramp must frequently adapt to the terrain conditions of different slopes, which presents a natural challenge to safety management.

[0003] Due to the varying slopes of ramps, workers are prone to slipping, falling, and other accidents due to loss of balance, slippery ground, or wind. Therefore, effective early warning and control can promptly identify unsafe behaviors and potential risks, reduce accidents, protect the lives of construction workers, ensure the smooth progress of the project, and improve the overall effectiveness of site safety management.

[0004] Existing safety monitoring generally adopts a uniform safety judgment standard, which ignores the dynamic impact of slope changes on safety risk thresholds. In steep slope areas, even a slight shift in a person's center of gravity can lead to a fall risk. If the lenient shift threshold in gentle slope areas is applied, the risk of steep slopes will be seriously underestimated. Conversely, an overly strict uniform standard in gentle slope areas may lead to unnecessary false alarms and warnings, resulting in a waste of regulatory resources. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, device and equipment for monitoring and early warning of personnel behavior at smart construction sites, which can monitor and provide early warning of personnel behavior at pumped storage power station construction sites.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] A method for monitoring and early warning of personnel behavior at smart construction sites includes:

[0008] Obtain complete terrain data of the target ramp in the target smart construction site;

[0009] Based on the complete terrain data, the target ramp is divided into multiple sub-regions with different slopes;

[0010] When the target person moves within any one of the multiple sub-regions, acquire the sequence of image and video frames of the target person moving within the sub-region within the sliding time window;

[0011] Obtain the center-of-gravity trajectory of the target person as they move within the image / video frame sequence and the slope of the sub-region where the target person is located within the sliding time window;

[0012] The initial behavioral safety index of the target person's actions is determined based on the center of gravity trajectory and slope.

[0013] Obtain wind speed time history data within the sub-region of the sliding time window;

[0014] Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person, calculate the wind-induced risk coefficient;

[0015] Obtain ground reflectance intensity data and environmental data within the sub-region of the sliding time window collected by multispectral sensors.

[0016] Calculate the wet slip risk coefficient of the sub-region within the slip time window based on the ground reflection intensity data and environmental data;

[0017] Based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, the comprehensive risk index of the target person moving within the sub-area is obtained;

[0018] The comprehensive risk index is used to issue early warnings about the behavior of the target personnel within the sub-region.

[0019] Optionally, based on the complete terrain data, the target ramp is divided into multiple sub-regions with different slopes, including:

[0020] Extract the longitudinal profile of the target ramp based on the complete terrain data;

[0021] The longitudinal profile line is divided into multiple equal-length foundation slope segments, and the average slope value of each equal-length foundation slope segment is calculated.

[0022] When the absolute value of the difference between the average slope values ​​of two adjacent equal-length foundation slope segments exceeds a preset threshold, the boundary point of the two adjacent equal-length foundation slope segments is determined as a forced boundary point.

[0023] The equal-length foundation slope segments between two adjacent forced boundary points are merged into a sub-region, resulting in multiple sub-regions.

[0024] Optionally, the initial behavioral safety index of the target person's actions is determined based on the center of gravity trajectory and slope, including:

[0025] The instantaneous offset is calculated based on the vertical distance between the center of gravity trajectory point of each frame and the projection line of the torso center.

[0026] The maximum offset amplitude of the center of gravity trajectory point within the sliding time window is determined based on the instantaneous offset.

[0027] The instantaneous offset is compared with a preset cumulative energy threshold to calculate the cumulative offset energy across multiple consecutive frames within the sliding time window; and...

[0028] The offset recovery rate of the target personnel is calculated based on the instantaneous offset.

[0029] The maximum offset amplitude, offset cumulative energy, and offset recovery rate are input into a preset fuzzy inference model for rule matching, and the initial behavioral safety index of the target personnel's behavior is output.

[0030] Optionally, based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the target person's posture, a wind-induced risk coefficient is calculated, including:

[0031] Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person, the equivalent overturning moment and the anti-overturning moment of the wind load acting on the human body are calculated respectively.

[0032] When the equivalent overturning moment exceeds the preset ratio of the anti-overturning moment, a wind-induced risk event is triggered, and the current wind-induced risk coefficient of the wind-induced risk event is calculated.

[0033] Obtain the current wind risk coefficient of all triggered wind-induced risk events within the sliding time window, and take the average of all triggered current wind risk coefficients to obtain the wind risk coefficient.

[0034] Optionally, the wet slip risk coefficient of the sub-region within the slip time window is calculated based on the ground reflection intensity data and environmental data, including:

[0035] The current slipperiness anomaly is calculated based on the ground reflection intensity data and the dry baseline value of the same sub-region at the same time in history;

[0036] The ground moisture evaporation attenuation function is calculated using the environmental data, and the current slipperiness anomaly is corrected using the ground moisture evaporation attenuation function to obtain the corrected anomaly.

[0037] The slippery risk coefficient is obtained based on the corrected anomaly degree.

[0038] Optionally, based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, a comprehensive risk index is obtained when the target person moves within the sub-area, including:

[0039] Through formula The comprehensive risk index of the target personnel when moving within the sub-region is obtained, wherein, For comprehensive risk index, Weights for the stability of the personnel's own center of gravity For the sake of wind and weight, For wet and slippery weight, For the initial behavioral safety index, For wind-induced risk coefficient, For slippery risk factor, The coupling coefficient;

[0040] Among them, the , and It varies with the slope of the sub-region.

[0041] Optional methods for monitoring and providing early warning of personnel behavior at smart construction sites also include:

[0042] Calculate the maximum allowable offset of the sub-region based on the slope of the sub-region where the target person is located within the sliding time window;

[0043] The instantaneous offset of each centroid trajectory point in the centroid trajectory is compared with the maximum allowable offset of the sub-region. When the instantaneous offset is greater than the maximum allowable offset, a high-risk warning is directly output.

[0044] The present invention also provides a smart construction site personnel behavior monitoring and early warning device, comprising:

[0045] The acquisition module is used to acquire complete terrain data of the target ramp in the target smart construction site;

[0046] The processing module is used to divide the target slope into multiple sub-regions with different slopes based on the complete terrain data; when a target person moves in any of the sub-regions, it acquires a sequence of image and video frames showing the target person moving in that sub-region within a sliding time window; it acquires the center-of-gravity trajectory of the target person and the slope of the sub-region where the target person is located within the sliding time window within the image and video frame sequence; it determines an initial behavioral safety index for the target person's actions based on the center-of-gravity trajectory and slope; it acquires wind speed time history data within the sub-region within the sliding time window; it calculates a wind-induced risk coefficient based on the wind speed time history data, the center-of-gravity trajectory of the target person, and the target person's posture; it acquires ground reflection intensity data collected by multispectral sensors and environmental data within the sub-region within the sliding time window; it calculates a slippery risk coefficient for the sub-region within the sliding time window based on the ground reflection intensity data and environmental data; it obtains a comprehensive risk index for the target person moving in the sub-region based on the initial behavioral safety index, the wind-induced risk coefficient, and the slippery risk coefficient; and it issues an early warning for the target person's actions within the sub-region based on the comprehensive risk index.

[0047] The present invention also provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above.

[0048] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above.

[0049] The above-described solution of the present invention has at least the following beneficial effects:

[0050] The above-described solution of the present invention acquires complete terrain data of a target ramp in a target smart construction site; divides the target ramp into multiple sub-regions with different slopes based on the complete terrain data; when a target person moves within any of the sub-regions, acquires a sequence of image and video frames showing the target person moving within that sub-region within a sliding time window; acquires the center of gravity trajectory of the target person during movement within the image and video frame sequence and the slope of the sub-region where the target person is located within the sliding time window; determines the initial behavioral safety index of the target person's actions based on the center of gravity trajectory and slope; acquires wind speed time history data within the sub-region within the sliding time window; and based on the wind speed time history data... The system calculates the wind-induced risk coefficient based on speed and time-course data, the center-of-gravity trajectory of the target personnel during movement, and the personnel's posture; it acquires ground reflection intensity data and environmental data within the sub-region collected by multispectral sensors within the sliding time window; it calculates the wet-slip risk coefficient of the sub-region within the sliding time window based on the ground reflection intensity data and environmental data; it obtains the comprehensive risk index of the target personnel moving within the sub-region based on the initial behavioral safety index, the wind-induced risk coefficient, and the wet-slip risk coefficient; and it provides early warning for the target personnel's behavior within the sub-region based on the comprehensive risk index, enabling monitoring and early warning of personnel behavior at pumped storage power station construction sites. Attached Figure Description

[0051] Figure 1 This is a flowchart of a smart construction site personnel behavior monitoring and early warning method according to an embodiment of the present invention;

[0052] Figure 2 This is a flowchart illustrating the initial behavior safety index calculation process of the smart construction site personnel behavior monitoring and early warning method according to an embodiment of the present invention.

[0053] Figure 3 This is a flowchart illustrating the early warning process of the smart construction site personnel behavior monitoring and early warning method according to an embodiment of the present invention.

[0054] Figure 4 This is a structural diagram of the smart construction site personnel behavior monitoring and early warning device according to an embodiment of the present invention. Detailed Implementation

[0055] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0056] like Figure 1As shown, an embodiment of the present invention proposes a method for monitoring and early warning of personnel behavior at smart construction sites, comprising:

[0057] Step 11: Obtain complete terrain data of the target ramp in the target smart construction site;

[0058] Here, the target ramp is a construction passage between a pumping station in a low-lying area and a reservoir at a higher elevation in a smart construction site, through which construction workers move back and forth. The complete terrain data is digital elevation data reflecting the shape of the ramp and its surrounding area, obtained through remote sensing mapping technology and processed by data calculation, point cloud filtering, and classification.

[0059] Step 12: Based on the complete terrain data, divide the target ramp into multiple sub-regions with different slopes;

[0060] A longitudinal section is obtained by longitudinally segmenting the digital elevation model generated based on the complete terrain data. The target slope follows the terrain changes, with its internal slope fluctuating up and down, and the longitudinal section is a winding curve with different slopes. In this invention, different safety benchmarks are configured for slopes with different slopes, and the longitudinal section is discretized, dividing it into multiple sub-regions with different slopes according to the slope.

[0061] Step 13: When the target person moves in any one of the multiple sub-regions, acquire the image and video frame sequence of the target person moving in the sub-region within the sliding time window;

[0062] Here, when the target person moves within the sub-area, the camera group in that sub-area is activated, and the corresponding video image is retrieved. This invention performs scrolling calculations and comprehensive judgments in units of sliding time windows. Preferably, the sliding time window duration is 60 seconds, and the sliding step size is 10 seconds. The sliding time window slides backward once every 10 seconds, and the system comprehensively processes all data collected in the current window.

[0063] Step 14: Obtain the trajectory of the center of gravity of the target person as they move within the image / video frame sequence and the slope of the sub-region where the target person is located within the sliding time window;

[0064] Here, for the image and video frame sequence images captured by the camera group, the target detection algorithm is used to identify the torso region of the target person in the image, and a rectangular bounding box containing the torso is output, specifying the coordinate parameters of the rectangular bounding box, such as the pixel coordinates of the upper left and lower right corners.

[0065] Based on the coordinates of the torso's rectangular bounding box in each frame, the geometric center of the rectangular bounding box is calculated. For example, the pixel coordinates of the center point are determined by the average of the coordinates of the upper left and lower right corners of the bounding box. This geometric center is then used as the center of gravity trajectory point of the target person's torso in the current frame.

[0066] The center-of-gravity trajectory points of each frame are connected sequentially in chronological order to form a cross-frame motion trajectory line of the target person in the image and video frame sequence, thus forming the center-of-gravity trajectory.

[0067] Step 15: Determine the initial behavioral safety index of the target person's actions based on the center of gravity trajectory and slope;

[0068] The initial behavioral safety index reflects the risk of imbalance in the actions of the target person when moving in the sub-region. In this invention, the safety threshold is dynamically adjusted by slope (tightening on steep slopes and widening on gentle slopes), and multiple risk features (such as maximum offset, cumulative energy, and offset recovery rate) are extracted from the center of gravity trajectory. The posture stability of the person is comprehensively quantified through fuzzy inference.

[0069] Step 16: Obtain wind speed time history data within the sub-region of the sliding time window;

[0070] Here, wind speed sensors deployed in the sub-region continuously collect the instantaneous wind speed values ​​of the sub-region at a high-frequency sampling period during the duration of the window (e.g., 60 seconds), forming a wind speed time series array arranged in chronological order.

[0071] Step 17: Calculate the wind-induced risk coefficient based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person;

[0072] The wind-induced risk coefficient reflects the degree to which wind force interferes with the balance of the target personnel.

[0073] Step 18: Obtain ground reflectance intensity data and environmental data within the sub-region of the sliding time window collected by the multispectral sensor;

[0074] Here, multispectral sensors and temperature and humidity sensors are pre-deployed in the sub-region. During the window period, ground multi-band reflection intensity data are collected synchronously at a fixed sampling period (e.g., every 10 seconds), and the current ambient temperature and relative humidity values ​​are obtained simultaneously.

[0075] Step 19: Calculate the wet slip risk coefficient of the sub-region within the slip time window based on the ground reflection intensity data and environmental data;

[0076] The slippery risk coefficient reflects the degree of threat that the slipperiness of the ground poses to the safety of pedestrians.

[0077] Step 110: Based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, obtain the comprehensive risk index of the target person when moving within the sub-area;

[0078] Here, the risks from three aspects are coupled together, and the influence of both the behavior itself and environmental factors is comprehensively considered to obtain a comprehensive risk index. This comprehensive risk index serves as the final quantitative evaluation of the safety status of personnel within this window, and its value ranges from 0 to 1.

[0079] Step 111: Issue an early warning regarding the target person's behavior within the sub-region based on the comprehensive risk index.

[0080] Here, the comprehensive risk index is mapped to different warning intervals, outputting three levels of warning signals: low risk, medium risk, and high risk, and simultaneously outputting the dominant risk factors. Preferably, the intervals are divided into: , , When the comprehensive risk index belongs to When the range is reached, a low-risk warning is output; when the comprehensive risk index belongs to... When the range is reached, a medium-risk warning is output; when the comprehensive risk index belongs to... When the interval is reached, a high-risk warning is issued.

[0081] This invention achieves precise personnel location and environmental adaptation by dividing complex slopes into sub-regions according to their slope and configuring sensing devices. The system uses a sliding time window as the evaluation unit, generating an initial behavioral safety index characterizing personnel stability based on a safety threshold derived from center-of-gravity trajectory extraction and dynamic slope correction. Simultaneously, it calculates the wind-induced overturning moment ratio using wind speed time history and body posture parameters to obtain a wind-induced risk coefficient. Furthermore, it obtains a slippery risk coefficient based on ground multispectral reflectance intensity and environmental temperature and humidity data, corrected for evaporation attenuation. Finally, it outputs a graded early warning signal by fusing the three coefficients through slope-adaptive dynamic weighting. This invention achieves a comprehensive evaluation from "single point in time" to "time window," from "uniform standard" to "slope adaptation," and from "behavioral isolation" to "wind-humidity-posture coupling," effectively solving the problems of ambiguous positioning, threshold mismatch, and high false alarm rates in traditional early warning systems, significantly improving the accuracy and reliability of safety management in complex terrain construction scenarios.

[0082] In an optional embodiment of the present invention, step 12, dividing the target slope into multiple sub-regions with different slopes based on the complete terrain data, may include:

[0083] Step 121: Extract the longitudinal profile of the target slope based on the complete terrain data;

[0084] Step 122: Divide the longitudinal profile line into multiple equal-length foundation slope segments and calculate the average slope value of each equal-length foundation slope segment.

[0085] Step 123: When the absolute value of the difference between the average slope values ​​of two adjacent equal-length foundation slope segments exceeds a preset threshold, the boundary point of the two adjacent equal-length foundation slope segments is determined as a forced boundary point.

[0086] Step 124: Merge the equal-length foundation slope segments between two adjacent forced boundary points into a sub-region, resulting in multiple sub-regions.

[0087] In this embodiment, a fixed length preset by the system is set as the standard length of the equal-length foundation slope segment based on the total length of the ramp and the accuracy requirements. Taking the starting point of the longitudinal profile line of the ramp as the reference, the ramp is cut segment by segment along the profile line according to the standard length of the equal-length foundation slope segment to generate continuous equal-length foundation slope segments. If the length of the remaining slope segment is less than the standard length of the equal-length foundation slope segment, it is marked as an equal-length foundation slope segment independently.

[0088] For each equal-length foundation slope segment, obtain the corresponding height values ​​of the starting point and the ending point, and calculate the height difference between the ending point and the starting point. The horizontal projection distance of the equal-length foundation slope segment is always the preset length, and the slope is defined as the percentage of the ratio of the height difference to the horizontal distance.

[0089] If the absolute value of the difference between the average slope values ​​of adjacent equal-length basic slope segments exceeds the preset slope threshold, the junction of the two slope segments will be marked as a forced boundary point; otherwise, the two slope segments will be merged into the same candidate sub-region.

[0090] It should be noted that the preset slope threshold is based on the safety regulations for pumped storage construction sites, specifically the safety standards for the sensitivity of slope changes during slope construction or personnel operations.

[0091] When the slope difference between adjacent equal-length base slope sections exceeds a threshold, i.e., a sudden change in slope, it is marked as a forced boundary point. This can separate areas with significantly different slope characteristics, so that each sub-area corresponds to a unique risk level. For example, the requirements for the stability of the center of gravity of personnel are different for steep slope sections and gentle slope sections, thus achieving precise control by slope zoning.

[0092] Merging adjacent slope segments of equal length with small slope differences can make the slope change within each sub-region gradual, avoiding overly detailed and redundant sub-region divisions due to frequent small fluctuations in slope, and ensuring that the identified slope angle of each sub-region can truly reflect its overall slope characteristics.

[0093] The profile is divided into multiple sub-regions using all forced boundary points as dividing points, and the average slope value of all equal-length foundation slope segments in each sub-region is used as the slope value of the sub-region.

[0094] In an optional embodiment of the present invention, such as Figure 2 As shown, in step 15, determining the initial behavioral safety index of the target person's actions based on the center of gravity trajectory and slope may include:

[0095] Step 151: Calculate the instantaneous offset based on the vertical distance between the center of gravity trajectory point of each frame in the center of gravity trajectory and the projection line of the torso center.

[0096] Here, the vertical projection line from the target person's center of gravity trajectory point to the target ramp when standing is taken as the static baseline. The vertical distance of each center of gravity trajectory point relative to the static baseline is calculated point by point as the instantaneous offset of the target person's center of gravity. For the i-th frame of the image / video frame sequence, the instantaneous offset of the target person's center of gravity in the i-th frame is denoted as... .

[0097] Step 152: Determine the maximum offset amplitude of the center of gravity trajectory point within the sliding time window based on the instantaneous offset.

[0098] The maximum offset value is expressed as This refers to the maximum instantaneous offset in the image / video frame sequence.

[0099] Step 153: Compare the instantaneous offset with the preset cumulative energy start threshold, and calculate the offset cumulative energy of multiple consecutive frames within the sliding time window;

[0100] Here, when the instantaneous offset exceeds the preset accumulated energy start threshold, it is recorded as one instance of offset energy accumulation. Whenever... hour, To preset the accumulated energy threshold, the accumulated offset energy is incremented once, resulting in a total accumulated offset energy within the final sliding time window. ,in The cumulative energy is defined as the offset, and N is the number of frames within the sliding time window where the instantaneous offset exceeds a preset cumulative energy threshold. The cumulative energy threshold is... Used to measure whether the target person is continuously shaking; when it exceeds... Representatives begin to expend attention to maintain balance, preferably... , This represents the maximum allowed offset for the sub-region. ,in The baseline offset is the safe radius of human body center of gravity swing measured through biomechanical experiments under a standard reference slope (usually defined as 0° horizontal ground or a very gentle slope). For empirical constants, Sub-region slope. (This refers to the slope of a sub-region.) The maximum allowable offset is modified to achieve the technical effect of tightening the threshold in steep slope sections and relaxing the threshold in gentle slope sections.

[0101] Step 154: Calculate the offset recovery rate of the target personnel based on the instantaneous offset.

[0102] The offset recovery rate reflects the ability of the target person to autonomously adjust their posture and restore balance. The offset recovery rate is denoted as... , The lower the value, the higher the risk. The offset recovery rate is calculated using the average negative rate of change of the center of gravity offset within a sliding time window.

[0103] First, for the instantaneous offset sequence arranged in chronological order within the sliding time window Calculate the rate of change of each frame relative to the previous frame: , Where n is the total number of frames within the sliding time window, For the timestamp of the i-th frame, For the (i-1)th frame timestamp, The instantaneous offset of the (i-1)th frame. The rate of change.

[0104] extract The interval, that is, the process in which the offset is decreasing, indicates that the personnel are actively adjusting their center of gravity back, ignoring... This means that the offset data points are being expanded.

[0105] Calculate the offset recovery rate ,in For weighting coefficients, The number of inter-frame changes in which the instantaneous shift of the centroid decreases within a sliding time window T, i.e., satisfying... The total number of sample points. If A high value indicates that once a person senses imbalance, they can quickly pull their center of gravity back to a safe zone, representing "strong posture control ability and low risk"; if... A value close to 0 indicates that the personnel have been in a state of deviation for a long time and are unable to correct their course, representing "fatigue or loss of control, and a sharp increase in risk".

[0106] Step 155: Input the maximum offset amplitude, offset cumulative energy and offset recovery rate into the preset fuzzy inference model for rule matching, and output the initial behavioral safety index of the target personnel's behavior.

[0107] Here, will , and Input a preset fuzzy inference model, in which a triangular membership function is defined, and... , and The mapping is represented by three fuzzy sets: "low," "medium," and "high." Based on the activated rules matched from the expert rule base, the risk level and the corresponding rule trigger strength are output. The expert rule base adopts the "IF-Then" format, for example: if... For high and If it is low, then output high risk; if For the middle and If the risk level is high, then the output is medium risk. Finally, the initial behavioral safety index is defuzzified using the centroid method to output a continuous index. . Where K is the number of discrete sampling points in the output universe of discourse. This is the membership envelope formed by aggregating all activation rules over the output universe.

[0108] In an optional embodiment of the present invention, step 17, calculating the wind-induced risk coefficient based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the target person's posture, may include:

[0109] Step 171: Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person, calculate the equivalent overturning moment and the anti-overturning moment of the wind load acting on the human body, respectively.

[0110] The equivalent overturning moment is expressed as: ,in To estimate the windward area based on the rectangular bounding box of the target personnel and the torso image, For the height of the human body's center of mass, The instantaneous dynamic pressure is calculated based on the current wind speed in the wind speed time history data. ,in For air density, Let be the wind speed at the current time t.

[0111] The anti-overturning moment is the restoring moment of the human body's own weight resisting external horizontal thrust (wind force) that causes it to roll around the edge of the foot. The anti-overturning moment is expressed as... ,in, To estimate the target personnel quality based on video images, , To effectively resist overturning lever arm.

[0112] Step 172: When the equivalent overturning moment exceeds the preset ratio of the anti-overturning moment, a wind-induced risk event is triggered, and the current wind-induced risk coefficient of the wind-induced risk event is calculated.

[0113] Here, the system calculates the ratio in real time at each sampling moment. Preferred, when A wind-induced risk event is triggered at this time. Current wind-induced risk coefficient. ,in, , The maximum wind speed in the past 3 seconds The average wind speed over the past 10 minutes, For gust factors. ,in For wind speed standard deviation, For average wind speed, The turbulence intensity is denoted as turbulence intensity. The preferred threshold is the critical gust factor. . The critical turbulence intensity threshold is preferred. .

[0114] Step 173: Obtain the current wind risk coefficient of all triggered wind-induced risk events within the sliding time window, and take the average of all triggered current wind risk coefficients to obtain the wind risk coefficient.

[0115] In an optional embodiment of the present invention, step 19, calculating the slippery risk coefficient of the sub-region within the slip time window based on the ground reflection intensity data and environmental data, may include:

[0116] Step 191: Calculate the current slipperiness anomaly based on the ground reflection intensity data and the dry baseline value of the same sub-region at the same time in history;

[0117] The dryness baseline value is taken as the average spectral reflectance of the same sub-region at the same time within several consecutive non-rainfall periods in the past. Through the formula Calculate the current slipperiness anomaly, where For ground reflectance intensity data collected by multispectral sensors, The current slipperiness level is abnormal.

[0118] Step 192: Calculate the ground moisture evaporation attenuation function using the environmental data, and correct the current slipperiness anomaly using the ground moisture evaporation attenuation function to obtain the corrected anomaly.

[0119] Here, the environmental data includes: ambient temperature. and ambient humidity The ground moisture evaporation attenuation function is: , where k is the evaporation rate coefficient and t is the cumulative time elapsed since the last significant slippery event. Corrected anomaly degree .

[0120] Step 193: Obtain the slippery risk coefficient based on the corrected anomaly degree.

[0121] The slippery risk factor ,in This is the preset critical anomaly level for slip risk.

[0122] In an optional embodiment of the present invention, such as Figure 3 As shown, in step 110, based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, the comprehensive risk index of the target person moving within the sub-area can be obtained, which may include:

[0123] Through formula The comprehensive risk index of the target personnel when moving within the sub-region is obtained, wherein, For comprehensive risk index, Weights for the stability of the personnel's own center of gravity For the sake of wind and weight, For wet and slippery weight, For the initial behavioral safety index, For wind-induced risk coefficient, For slippery risk factor, The coupling coefficient;

[0124] Among them, the , and It varies with the slope of the sub-region.

[0125] In this embodiment, a dynamic risk weight is set that follows the slope changes of the sub-region where the target person is located. When the sub-region has a large slope, i.e., it is in a steep slope area, the weight of the person's own center of gravity stability is adjusted. Increase, wind weight The wet slip weight is also increased due to the wind effect at higher altitudes. When the sub-region has a small slope, i.e., it is in a gentle slope area, the wet slip weight is increased. Increased slipperiness is a dominant factor as gentle slopes are easier to walk on, people become less vigilant, and slipperiness becomes the primary factor.

[0126] Preferred, , , .

[0127] In an optional embodiment of the present invention, such as Figure 2 As shown, the methods for monitoring and early warning of personnel behavior in smart construction sites also include:

[0128] Step 21: Calculate the maximum allowable offset of the sub-region based on the slope of the sub-region where the target person is located within the sliding time window;

[0129] Maximum allowable offset ,in For the reference offset, For empirical constants, The slope of the sub-region.

[0130] Step 22: Compare the instantaneous offset of each center of gravity trajectory point in the center of gravity trajectory with the maximum allowable offset of the sub-region. When the instantaneous offset is greater than the maximum allowable offset, output a high-risk warning directly.

[0131] In this embodiment, an interruption warning mechanism is set up. When an interruption warning is detected... When this occurs, an interruption warning will be immediately triggered, and an emergency high-risk warning will be output.

[0132] It should be noted that in the above embodiments of the present invention, when the target person moves across sub-regions, for example from a slope... Area a moves to the slope In region b, the calculation of the sliding time window across sub-regions employs a logic where the time window is not reset, but the spatial reference is dynamically anchored. That is, the sliding time window does not stop when crossing sub-regions, and the spatial reference is based on the slope of the sub-region where the sliding time window ends.

[0133] Specifically, for the calculation of the initial behavioral safety index, it is assumed that the target personnel are moving from a slope... Area a moves to the slope In region b, the maximum allowable offset is... according to Calculate the corresponding cumulative energy threshold. for The corresponding cumulative energy start-up threshold. The instantaneous offset acquired in each frame within the sliding time window is... All with The corresponding cumulative energy threshold is compared. Simultaneously, the instantaneous offset acquired in each frame within the sliding time window is... Also with Corresponding maximum allowable offset When comparing, Greater than Corresponding maximum allowable offset If this occurs, an interruption warning will be triggered immediately.

[0134] Regarding the wind-induced risk coefficient, due to the strong time-varying nature of wind and its significant influence from terrain obstruction, wind data from older areas within a historical window becomes completely irrelevant in cross-regional scenarios. All wind speed sampling points collected in area a within the sliding time window are not included in the wind-induced risk coefficient calculation for that window. At the moment the window closes, the system directly reads the wind speed sensor data from area b, where the target person is located, obtaining the wind speed time-history data for the most recent 10 seconds. The gust factor is then extracted based on the wind speed time-history data for the most recent 10 seconds. and turbulence intensity Calculate the wind-induced risk coefficient.

[0135] Regarding the risk factor of slippery surfaces, slippery surfaces exhibit strong spatial memory and gradual change. If a person moves from area A to area B, and area A is extremely slippery, even though the person has left, the postural tension caused by the slipperiness is still reflected in their center of gravity trajectory. When the target person moves from a slope... Area a moves to the slope For region b, the system retrieves the corrected anomalies of regions a and b at the current sliding time window cutoff time. Take the corrected anomaly degree Substitute the maximum value in the formula Calculate the slippery risk factor.

[0136] The cross-sub-region sliding time window risk calculation method provided by this invention effectively solves the problem of warning jumps and inaccuracies at slope change boundaries in traditional methods by designing differentiated cross-regional fusion strategies for the initial behavioral safety index, wind-induced risk coefficient, and wet slip risk coefficient. For the initial behavioral safety index, a "slope locking at the cutoff time and historical frame backtracking and re-judgment" mechanism is adopted to re-evaluate all center-of-gravity data within the window with the dynamic safety threshold of the current region, avoiding missed or false alarms caused by slope switching; the wind-induced risk coefficient adopts the "field value replacement method" and is calculated only with the latest real-time wind speed in the region at the cutoff time of the window, eliminating the dilution of current risk by outdated wind data; the wet slip risk coefficient adopts the "maximum spatial footprint preservation method" and takes the maximum value of the wet slip index of all sub-regions traversed by personnel within the window, ensuring the continuity of the slowly changing risk of ground slippage during spatial transfer. The three strategies work together to enable the early warning system to respond sensitively to changes in slope and environment in new areas when people move across regions, while also reasonably inheriting risk information from historical periods, significantly improving the continuity and physical accuracy of safety early warnings in complex terrain.

[0137] like Figure 4 As shown, an embodiment of the present invention also provides a smart construction site personnel behavior monitoring and early warning device 40, comprising:

[0138] Module 41 is used to acquire complete terrain data of the target ramp in the target smart construction site;

[0139] Processing module 42 is used to divide the target slope into multiple sub-regions with different slopes based on the complete terrain data; when the target person moves in any of the multiple sub-regions, it acquires an image and video frame sequence of the target person moving in the sub-region within a sliding time window; acquires the center of gravity trajectory of the target person during movement within the image and video frame sequence and the slope of the sub-region where the target person is located within the sliding time window; determines the initial behavioral safety index of the target person's behavior based on the center of gravity trajectory and slope; acquires wind speed time history data within the sub-region within the sliding time window; and, based on the... The wind-induced risk coefficient is calculated based on wind speed time-history data, the center-of-gravity trajectory of the target person during movement, and the target person's posture. Ground reflection intensity data and environmental data within the sub-region, collected by multispectral sensors, are acquired within the sliding time window. A wet-slip risk coefficient for the sub-region within the sliding time window is calculated based on the ground reflection intensity data and environmental data. A comprehensive risk index is obtained based on the initial behavioral safety index, the wind-induced risk coefficient, and the wet-slip risk coefficient, allowing the target person to move within the sub-region. An early warning is issued for the target person's actions within the sub-region based on the comprehensive risk index.

[0140] Optionally, based on the complete terrain data, the target ramp is divided into multiple sub-regions with different slopes, including:

[0141] Extract the longitudinal profile of the target ramp based on the complete terrain data;

[0142] The longitudinal profile line is divided into multiple equal-length foundation slope segments, and the average slope value of each equal-length foundation slope segment is calculated.

[0143] When the absolute value of the difference between the average slope values ​​of two adjacent equal-length foundation slope segments exceeds a preset threshold, the boundary point of the two adjacent equal-length foundation slope segments is determined as a forced boundary point.

[0144] The equal-length foundation slope segments between two adjacent forced boundary points are merged into a sub-region, resulting in multiple sub-regions.

[0145] Optionally, the initial behavioral safety index of the target person's actions is determined based on the center of gravity trajectory and slope, including:

[0146] The instantaneous offset is calculated based on the vertical distance between the center of gravity trajectory point of each frame and the projection line of the torso center.

[0147] The maximum offset amplitude of the center of gravity trajectory point within the sliding time window is determined based on the instantaneous offset.

[0148] The instantaneous offset is compared with a preset cumulative energy threshold to calculate the cumulative offset energy across multiple consecutive frames within the sliding time window; and...

[0149] The offset recovery rate of the target personnel is calculated based on the instantaneous offset.

[0150] The maximum offset amplitude, offset cumulative energy, and offset recovery rate are input into a preset fuzzy inference model for rule matching, and the initial behavioral safety index of the target personnel's behavior is output.

[0151] Optionally, based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the target person's posture, a wind-induced risk coefficient is calculated, including:

[0152] Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person, the equivalent overturning moment and the anti-overturning moment of the wind load acting on the human body are calculated respectively.

[0153] When the equivalent overturning moment exceeds the preset ratio of the anti-overturning moment, a wind-induced risk event is triggered, and the current wind-induced risk coefficient of the wind-induced risk event is calculated.

[0154] Obtain the current wind risk coefficient of all triggered wind-induced risk events within the sliding time window, and take the average of all triggered current wind risk coefficients to obtain the wind risk coefficient.

[0155] Optionally, the wet slip risk coefficient of the sub-region within the slip time window is calculated based on the ground reflection intensity data and environmental data, including:

[0156] The current slipperiness anomaly is calculated based on the ground reflection intensity data and the dry baseline value of the same sub-region at the same time in history;

[0157] The ground moisture evaporation attenuation function is calculated using the environmental data, and the current slipperiness anomaly is corrected using the ground moisture evaporation attenuation function to obtain the corrected anomaly.

[0158] The slippery risk coefficient is obtained based on the corrected anomaly degree.

[0159] Optionally, based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, a comprehensive risk index is obtained when the target person moves within the sub-area, including:

[0160] Through formula The comprehensive risk index of the target personnel when moving within the sub-region is obtained, wherein, For comprehensive risk index, Weights for the stability of the personnel's own center of gravity For the sake of wind and weight, For wet and slippery weight, For the initial behavioral safety index, For wind-induced risk coefficient, For slippery risk factor, The coupling coefficient;

[0161] Among them, the , and It varies with the slope of the sub-region.

[0162] Optionally, the processing module 42 is also used for:

[0163] Calculate the maximum allowable offset of the sub-region based on the slope of the sub-region where the target person is located within the sliding time window;

[0164] The instantaneous offset of each centroid trajectory point in the centroid trajectory is compared with the maximum allowable offset of the sub-region. When the instantaneous offset is greater than the maximum allowable offset, a high-risk warning is directly output.

[0165] It should be noted that this device is the same as the method described above. All implementations in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effect.

[0166] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0167] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method as described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0168] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0169] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0170] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0171] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0172] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0173] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0174] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve by using their basic programming skills after reading the description of the present invention.

[0175] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0176] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for monitoring and early warning of personnel behavior at smart construction sites, characterized in that, include: Obtain complete terrain data of the target ramp in the target smart construction site; Based on the complete terrain data, the target ramp is divided into multiple sub-regions with different slopes; When the target person moves within any one of the multiple sub-regions, acquire the sequence of image and video frames of the target person moving within the sub-region within the sliding time window; Obtain the center-of-gravity trajectory of the target person as they move within the image / video frame sequence and the slope of the sub-region where the target person is located within the sliding time window; The initial behavioral safety index of the target person's actions is determined based on the center of gravity trajectory and slope. Obtain wind speed time history data within the sub-region of the sliding time window; Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person, calculate the wind-induced risk coefficient; Obtain ground reflectance intensity data and environmental data within the sub-region of the sliding time window collected by multispectral sensors. Calculate the wet slip risk coefficient of the sub-region within the slip time window based on the ground reflection intensity data and environmental data; Based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, the comprehensive risk index of the target person moving within the sub-area is obtained; The comprehensive risk index is used to issue early warnings about the behavior of the target personnel within the sub-region.

2. The method for monitoring and early warning of personnel behavior at smart construction sites according to claim 1, characterized in that, Based on the complete terrain data, the target ramp is divided into multiple sub-regions with different slopes, including: Extract the longitudinal profile of the target ramp based on the complete terrain data; The longitudinal profile line is divided into multiple equal-length foundation slope segments, and the average slope value of each equal-length foundation slope segment is calculated. When the absolute value of the difference between the average slope values ​​of two adjacent equal-length foundation slope segments exceeds a preset threshold, the boundary point of the two adjacent equal-length foundation slope segments is determined as a forced boundary point. The equal-length foundation slope segments between two adjacent forced boundary points are merged into a sub-region, resulting in multiple sub-regions.

3. The method for monitoring and early warning of personnel behavior at smart construction sites according to claim 1, characterized in that, The initial behavioral safety index of the target person's actions is determined based on the center of gravity trajectory and slope, including: The instantaneous offset is calculated based on the vertical distance between the center of gravity trajectory point of each frame and the projection line of the torso center. The maximum offset amplitude of the center of gravity trajectory point within the sliding time window is determined based on the instantaneous offset. The instantaneous offset is compared with a preset cumulative energy threshold to calculate the cumulative offset energy across multiple consecutive frames within the sliding time window; and... The offset recovery rate of the target personnel is calculated based on the instantaneous offset. The maximum offset amplitude, offset cumulative energy, and offset recovery rate are input into a preset fuzzy inference model for rule matching, and the initial behavioral safety index of the target personnel's behavior is output.

4. The method for monitoring and early warning of personnel behavior at smart construction sites according to claim 1, characterized in that, Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the target person's posture, the wind-induced risk coefficient is calculated, including: Based on the wind speed time history data, the center of gravity trajectory of the target person during movement, and the body posture of the target person, the equivalent overturning moment and the anti-overturning moment of the wind load acting on the human body are calculated respectively. When the equivalent overturning moment exceeds the preset ratio of the anti-overturning moment, a wind-induced risk event is triggered, and the current wind-induced risk coefficient of the wind-induced risk event is calculated. Obtain the current wind risk coefficient of all triggered wind-induced risk events within the sliding time window, and average the current wind risk coefficients of all triggered events to obtain the wind risk coefficient.

5. The method for monitoring and early warning of personnel behavior at smart construction sites according to claim 1, characterized in that, The wet slip risk coefficient of the sub-region within the slip time window is calculated based on the ground reflection intensity data and environmental data, including: The current slipperiness anomaly is calculated based on the ground reflection intensity data and the dry baseline value of the same sub-region at the same time in history; The ground moisture evaporation attenuation function is calculated using the environmental data, and the current slipperiness anomaly is corrected using the ground moisture evaporation attenuation function to obtain the corrected anomaly. The slippery risk coefficient is obtained based on the corrected anomaly degree.

6. The method for monitoring and early warning of personnel behavior at smart construction sites according to claim 1, characterized in that, Based on the initial behavioral safety index, wind-induced risk coefficient, and slippery risk coefficient, a comprehensive risk index is obtained when the target person moves within the sub-area, including: Through formula The comprehensive risk index of the target personnel when moving within the sub-region is obtained, wherein, For comprehensive risk index, Weights for the stability of the personnel's own center of gravity For the sake of wind and weight, For wet and slippery weight, For the initial behavioral safety index, For wind-induced risk coefficient, For slippery risk factor, The coupling coefficient; Among them, the , and It varies with the slope of the sub-region.

7. The method for monitoring and early warning of personnel behavior at smart construction sites according to claim 1, characterized in that, Also includes: Calculate the maximum allowable offset of the sub-region based on the slope of the sub-region where the target person is located within the sliding time window; The instantaneous offset of each centroid trajectory point in the centroid trajectory is compared with the maximum allowable offset of the sub-region. When the instantaneous offset is greater than the maximum allowable offset, a high-risk warning is directly output.

8. A smart construction site personnel behavior monitoring and early warning device, characterized in that, include: The acquisition module is used to acquire complete terrain data of the target slope in the target smart construction site; The processing module is used to divide the target slope into multiple sub-regions with different slopes based on the complete terrain data; when a target person moves in any of the sub-regions, it acquires a sequence of image and video frames showing the target person moving in that sub-region within a sliding time window; it acquires the center-of-gravity trajectory of the target person and the slope of the sub-region where the target person is located within the sliding time window within the image and video frame sequence; it determines the initial behavioral safety index of the target person's behavior based on the center-of-gravity trajectory and slope; it acquires wind speed time history data within the sub-region within the sliding time window; it calculates a wind-induced risk coefficient based on the wind speed time history data, the center-of-gravity trajectory of the target person, and the target person's posture; it acquires ground reflection intensity data collected by multispectral sensors and environmental data within the sub-region within the sliding time window; it calculates a slippery risk coefficient for the sub-region within the sliding time window based on the ground reflection intensity data and environmental data; and it obtains a comprehensive risk index for the target person moving in the sub-region based on the initial behavioral safety index, the wind-induced risk coefficient, and the slippery risk coefficient. The comprehensive risk index is used to issue early warnings about the behavior of the target personnel within the sub-region.

9. A computing device, characterized in that, include: A processor, a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A storage instruction that, when executed on a computer, causes the computer to perform the method as described in any one of claims 1 to 7.

Images