Firefighter posture monitoring method and system based on flexible bending sensor

Abstract

The invention discloses a fireman posture monitoring method and system based on a flexible bending sensor, and relates to the technical field of fireman posture monitoring methods, and the method comprises the steps: determining a bending change diagram of the flexible bending sensor based on the change of bending data of the flexible bending sensor; according to the bending change diagram, the action path of the fire-fighting action and the current attitude schematic diagram of the firefighter, the abnormal attitude node is determined, the monitoring accuracy of the abnormal attitude node is improved, and the corresponding abnormal action path is determined based on the bending data and the action position of the abnormal attitude node. Determining an abnormal area according to the plurality of abnormal action paths, the corresponding action environment and the current position of the firefighter; the risk action of the firefighter is determined based on the abnormal area and the corresponding fire action, and the posture correction measure of the firefighter is determined according to the risk action, the corresponding posture data and the bending data of the flexible bending sensor, so that the accuracy of the posture correction measure of the firefighter is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of firefighter posture monitoring methods, and in particular to a firefighter posture monitoring method and system based on a flexible bending sensor. Background Art

[0002] With the development of science and technology, firefighters are equipped with tights during training. The tights have built-in flexible bending sensors. Each flexible bending sensor is arranged relative to various parts of the firefighter and changes flexibly with the firefighter's movements to collect corresponding bending data. In the existing technology, the firefighter's posture is determined based on the identification of bending data. However, the abnormal posture nodes of the firefighter cannot be known, which affects the accuracy of the firefighter's posture correction measures, and thus the firefighter cannot achieve dynamic adjustment of risky actions. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a firefighter posture monitoring method and system based on a flexible bending sensor.

[0004] An embodiment of the present invention provides a firefighter posture monitoring method based on a flexible bending sensor, comprising: determining a firefighter's current posture schematic diagram based on the firefighter's current posture data and body shape data, and marking the bending data of the flexible bending sensor configured on the firefighter's clothes; when the firefighter performs a firefighting action, determining a bending change diagram of the flexible bending sensor based on changes in the bending data of the flexible bending sensor; determining an abnormal posture node based on the bending change diagram, the action path of the firefighting action and the firefighter's current posture schematic diagram; during the monitoring process of the abnormal posture node, determining a corresponding abnormal action path based on the bending data and action position of the abnormal posture node, and determining an abnormal area based on multiple abnormal action paths, corresponding action environments and the firefighter's current position; determining the firefighter's risky action based on the abnormal area and the corresponding firefighting action, and determining the firefighter's posture corrective measures based on the risky action, the corresponding posture data and the bending data of the flexible bending sensor, and triggering the firefighter to dynamically adjust the risky action.

[0005] An embodiment of the present invention provides a firefighter posture monitoring system based on a flexible bending sensor. The firefighter posture monitoring system based on a flexible bending sensor is applied to the above-mentioned firefighter posture monitoring method based on a flexible bending sensor. The firefighter posture monitoring system based on a flexible bending sensor includes: a posture diagram module, configured to determine a current posture diagram of the firefighter based on the current posture data and body shape data of the firefighter, and mark bending data of a flexible bending sensor disposed on the firefighter's clothes; a bending change map module for determining a bending change map of the flexible bending sensor based on changes in bending data of the flexible bending sensor when a firefighter performs a firefighting action; An abnormal posture node module is used to determine an abnormal posture node according to the bending change graph, the action path of the firefighting action and the current posture diagram of the firefighter; An abnormal area module is used to determine the corresponding abnormal motion path based on the bending data and motion position of the abnormal posture node during the monitoring process of the abnormal posture node, and determine the abnormal area according to multiple abnormal motion paths, corresponding motion environments and the current position of the firefighter; The posture corrective measures module is used to determine the firefighter's risky actions based on the abnormal area and the corresponding firefighting actions, and to determine the firefighter's posture corrective measures based on the risky actions, the corresponding posture data and the bending data of the flexible bending sensor, and to trigger the firefighter's dynamic adjustment of the risky actions.

[0006] Compared with the prior art, the present invention has the following beneficial effects: In an embodiment of the present invention, through the method in the embodiment of the present invention, when a firefighter performs a firefighting action, a bending change graph of the flexible bending sensor is determined based on the change of the bending data of the flexible bending sensor; the abnormal posture node is determined according to the bending change graph, the action path of the firefighting action and the firefighter's current posture schematic diagram, and the bending change graph of the flexible bending sensor is introduced, which is compatible with the overall consideration of the bending change graph, the action path of the firefighting action and the firefighter's current posture schematic diagram, thereby improving the monitoring accuracy of the abnormal posture node.

[0007] Therefore, in the monitoring process of the abnormal posture node, the corresponding abnormal action path is determined based on the bending data and action position of the abnormal posture node, and the abnormal area is determined according to multiple abnormal action paths, the corresponding action environment and the current position of the firefighter; the firefighter's risky action is determined based on the abnormal area and the corresponding firefighting action, and the firefighter's posture correction measures are determined according to the risky action, the corresponding posture data and the bending data of the flexible bending sensor, and the firefighter's dynamic adjustment of the risky action is triggered, thereby realizing further control of the risky action, and then realizing the overall consideration of the risky action, the corresponding posture data and the bending data of the flexible bending sensor, ensuring the accuracy of the firefighter's posture correction measures, and realizing the dynamic adjustment of the risky action by the firefighter. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 1 is a flow chart of a firefighter posture monitoring method based on a flexible bending sensor in an embodiment of the present invention; Figure 2 Schematic diagram of the structure of a firefighter posture monitoring system based on a flexible bending sensor in an embodiment of the present invention. DETAILED DESCRIPTION

[0009] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0010] See also Figure 1 and Figure 2 A firefighter posture monitoring method based on a flexible bending sensor is applied to a firefighter posture monitoring scenario. The firefighter posture monitoring method based on a flexible bending sensor includes: Step S11: determining a current posture diagram of the firefighter based on the current posture data and body shape data of the firefighter, and marking bending data of a flexible bending sensor configured on the firefighter's clothes; Step S12: determining a bending change graph of the flexible bending sensor based on changes in bending data of the flexible bending sensor when the firefighter performs a firefighting action; Step S13: determining abnormal posture nodes according to the bending change graph, the action path of the firefighting action, and the firefighter's current posture diagram; Step S14: During the monitoring process of the abnormal posture node, a corresponding abnormal motion path is determined based on the bending data and motion position of the abnormal posture node, and an abnormal area is determined based on the multiple abnormal motion paths, the corresponding motion environment, and the current position of the firefighter; Step S15: determining the risky action of the firefighter based on the abnormal area and the corresponding firefighting action, determining the firefighter's posture correction measures based on the risky action, the corresponding posture data, and the bending data of the flexible bending sensor, and triggering the firefighter to dynamically adjust the risky action; In step S11, a firefighter's current posture diagram is determined based on the firefighter's current posture data and body shape data, and bending data of a flexible bending sensor configured on the firefighter's clothes is marked; In the specific implementation process of the present invention, the specific steps are: S111: The current position of the firefighter is collected, a corresponding target detection area is determined based on the current position of the firefighter and surrounding cameras, current posture data of the firefighter is determined based on real-time detection of the target detection area, and a current posture diagram of the firefighter is constructed based on the current posture data, the corresponding time node, and the body contour image of the firefighter; S112: determining key posture parts of the firefighter based on the detection of the firefighter's current posture diagram, and determining the flexible bending sensor configured on the firefighter's clothes according to the position detection of the key posture parts; S113: real-time monitoring of the flexible bending sensor; determining bending data recorded by the flexible bending sensor as the firefighter's posture changes based on the real-time monitoring of the flexible bending sensor, the bending data showing the bending degree of the flexible bending sensor and the degree of change in the firefighter's posture.

[0011] In an embodiment of the present application, the current location of the firefighter is collected, and a positioning technology suitable for a complex environment (such as an indoor fire scene) is selected. Common positioning technologies include GPS, UWB (ultra-wideband), Bluetooth beacons, etc. In indoor environments, the GPS signal may be weak, so UWB or Bluetooth beacons may be a better choice. A positioning module is installed on the firefighter's equipment (such as a fire helmet, fire suit or belt), and a positioning base station (such as a UWB base station or Bluetooth beacon) is deployed at the fire scene to receive and process the positioning signal. The positioning module collects the firefighter's location data in real time and transmits the data to the monitoring system through wireless communication (such as Bluetooth or Wi-Fi). The monitoring system receives the location data and stores it in a database.

[0012] Multiple cameras are deployed at the fire scene to ensure coverage of key areas (such as entrances, corridors, stairways, etc.). The cameras can be fixed cameras or mobile cameras (such as cameras mounted on drones). Based on the current location of the firefighter, the nearest camera is selected as the target detection device to determine the scope of the target detection area to ensure that the firefighter's full body posture can be fully captured. The camera collects the firefighter's image data in real time and transmits the data to the monitoring system via a wired or wireless network. The monitoring system processes the received image data and extracts the firefighter's posture information.

[0013] The image data collected by the camera is preprocessed, including denoising and contrast enhancement, to improve image quality. Background interference is removed using background subtraction or frame difference methods to highlight the outline of the firefighter. The processed image is then subjected to posture detection using deep learning algorithms (such as OpenPose and YOLO). The algorithm can identify the position and angle of key parts of the firefighter's body (such as the head, shoulders, elbows, and knees). The firefighter's current posture data, including the position coordinates and angle information of each key part, is extracted from the detection results and stored in the monitoring system's database for subsequent analysis.

[0014] While collecting posture data, the corresponding time nodes are recorded to ensure the synchronization of posture data and time. The time nodes can be used for subsequent posture change analysis and historical data backtracking. The firefighter's body contour image is extracted from the image captured by the camera. The body contour is extracted using an edge detection algorithm (such as Canny edge detection) or a deep learning algorithm (such as semantic segmentation). The posture data, time nodes and body contour image are combined to construct a schematic diagram of the firefighter's current posture. The posture diagram can be a three-dimensional model or a two-dimensional image, which intuitively displays the firefighter's body posture and the relative positions of various parts.

[0015] Alternatively, assume that at a large warehouse fire scene, firefighters enter the building to conduct firefighting and search and rescue operations. A UWB positioning module is installed on the firefighter's helmet, and multiple UWB base stations are deployed around the scene. The UWB positioning module collects the firefighter's location data every 0.1 seconds and transmits the data to the monitoring system via Bluetooth. After receiving this location data, the monitoring system updates the firefighter's location information in real time. After the firefighter enters the building, the monitoring system selects the nearest fixed camera as the target detection device based on the firefighter's UWB positioning data. This camera covers the corridor area where the firefighter is located. The camera collects image data every 0.03 seconds and transmits the data to the monitoring system via a wired network. The monitoring system processes the received image data in real time and extracts information about the firefighter's posture.

[0016] After receiving image data from the camera, the monitoring system first performs image denoising and contrast enhancement. Then, the OpenPose algorithm is used to detect poses in the processed images. The OpenPose algorithm identifies the position coordinates and angles of key body parts of the firefighter, such as the head, shoulders, elbows, and knees. For example, the detection results show that the firefighter's left shoulder is at (x1, y1), the right shoulder is at (x2, y2), the left elbow angle is 30°, and the right elbow angle is 45°. This pose data is stored in the monitoring system's database. The system records the firefighter's pose data along with the corresponding time (e.g., June 28, 2025, 14:30:00). The Canny edge detection algorithm is used to extract the firefighter's body outline from the camera image. Then, a two-dimensional pose diagram is constructed by combining the pose data (e.g., left shoulder position (x1, y1), right shoulder position (x2, y2), left elbow angle 30°, right elbow angle 45°) with the body outline image. The posture diagram shows the firefighter currently in a crouched, forward-crawl position, with the left arm bent at 30 degrees and the right arm bent at 45 degrees. The monitoring system displays this posture diagram in real time on the command center's screen, allowing command personnel to quickly understand the firefighter's status.

[0017] Furthermore, based on the detection of the firefighter's current posture diagram, the key parts of the firefighter's posture are determined, and according to the position detection of the key parts of the posture, the flexible bending sensor configured on the firefighter's clothes is determined, which is compatible with the overall consideration of the position detection of the key parts of the posture and ensures the accuracy of the flexible bending sensor configured on the firefighter's clothes.

[0018] Key posture parts are those that change most significantly during a firefighter's movements and are most helpful for posture assessment. These typically include the shoulders, elbows, wrists, waist, hips, knees, and ankles. The bending angles and position changes of these parts can reflect a firefighter's primary movements, such as crawling, squatting, bending, and lifting. By analyzing the firefighter's current posture diagram, the positions and angles of these key posture parts can be identified.

[0019] The installation location of the flexible bending sensor should be determined based on the location of key posture parts. The sensor should be installed in a position that can accurately capture the bending changes of key parts. For example, for the elbows and knees, the sensor can be installed on the sleeves and trouser legs of the firefighter uniform; for the waist, the sensor can be installed at the waist position of the firefighter uniform. Design a suitable fixing device to ensure that the sensor does not loosen or shift during the firefighter's movement. The sensor can be fixed to the firefighter uniform using elastic bandages, Velcro, or built-in pockets.

[0020] After the firefighters don their gear, they perform an initial sensor calibration. Calibrate the sensor using known bending angles (e.g., 0° for a straight arm, 90° for a 90° bend) to ensure the accuracy of the measurement data. Simple calibration movements (e.g., straight arm, 90° bend, knee straight, 90° bend) can be used for calibration. The sensor's response is tested as the firefighters perform basic movements (e.g., raising their hands, bending over, squatting) to ensure it accurately captures bending changes. If the sensor's response is inaccurate, adjust the sensor's position or recalibrate.

[0021] Alternatively, assume a firefighter is performing a ladder climbing task. The monitoring system uses a camera and posture detection algorithm to construct a diagram of the firefighter's current posture. This diagram shows the firefighter's left shoulder at (x1, y1), right shoulder at (x2, y2), left elbow angle at 30°, right elbow angle at 45°, waist flexion angle at 15°, left knee angle at 90°, and right knee angle at 100°. By analyzing this data, the key posture components can be identified as the left elbow, right elbow, waist, and knees, as changes in the flexion angles of these components are most critical for determining ladder climbing performance.

[0022] During the aforementioned ladder-climbing task, the monitoring system determined that the critical posture areas are the left elbow, right elbow, waist, and knees. Based on the locations of these areas, the following flexible bending sensors were installed on the firefighters' gear: Left elbow sensor: Installed at the elbow of the left sleeve of the firefighter's suit. Right elbow sensor: Installed at the elbow of the right sleeve of the firefighter's suit. Waist sensor: Installed at the waist of the firefighter's suit, close to the spine. Left knee sensor: Installed at the knee of the left trouser leg of the firefighter's suit. Right knee sensor: Installed at the knee of the right trouser leg of the firefighter's suit. These sensors are fixed to the firefighter's suit with elastic bandages to ensure that bending changes can be stably captured during the firefighter's ladder climbing process.

[0023] After the firefighter dons their gear and installs the sensors, an initial calibration is performed. The firefighter straightens their arm, and the monitoring system records an initial value of 0° for the left and right elbow sensors. Then, the firefighter bends their arm 90°, and the monitoring system records a value of 90° for the left and right elbow sensors. This calibrates all sensors. Next, the firefighter performs some basic movement tests. For example, when the firefighter bends, the waist sensor records a 30° bend angle. When the firefighter squats, the knee sensors record bend angles of 90° and 100°, respectively. The monitoring system confirms that the sensor responses are accurate and that the sensor installation positions perfectly match the key posture areas.

[0024] Therefore, the flexible bending sensor is monitored in real time; the bending data recorded by the flexible bending sensor as the firefighter's posture changes is determined based on the real-time monitoring of the flexible bending sensor. The bending data shows the degree of bending of the flexible bending sensor and the degree of change of the firefighter's posture, and is compatible with the overall consideration of the real-time monitoring of the flexible bending sensor, ensuring the accuracy of the bending data recorded by the flexible bending sensor as the firefighter's posture changes.

[0025] At this time, after the firefighters put on the equipment and complete the sensor calibration, the real-time monitoring function of the sensor is activated to ensure that the sensor is connected to the monitoring system (such as the server or mobile terminal of the command center) through wireless communication (such as Bluetooth, Wi-Fi) or wired communication. The sensor needs to have the characteristics of low power consumption, high sampling rate and low latency to ensure the real-time and accuracy of the data. The sensor collects bending data in real time at a set sampling rate (such as 50Hz or 100Hz). The bending data includes information such as the sensor's bending angle, bending speed and bending direction. The collected data is transmitted to the monitoring system in real time through the wireless communication module.

[0026] The monitoring system processes the received bending data in real time, including filtering, denoising and data correction, and extracts key features of the bending data, such as the range of change of the bending angle, the peak value of the bending speed, etc. The processed data is combined with the firefighter's current posture diagram to analyze the firefighter's posture changes. The processed bending data is stored in the database of the monitoring system, and the data timestamp, sensor position and bending status are recorded. The data storage format can be in tabular form to facilitate subsequent analysis and backtracking.

[0027] The processed bending data is displayed in real time on the interface of the monitoring system to present the changes in the firefighter's posture in an intuitive manner. Charts (such as line charts and bar charts) or animations (such as dynamic displays of three-dimensional models) can be used to display the data. The degree of change in the firefighter's posture can be judged based on the range and speed of change of the bending data. For example, a rapid change in the bending angle may indicate that the firefighter is performing a violent action, such as climbing a ladder quickly or avoiding danger, while a slow change in the bending angle may indicate that the firefighter is performing a relatively stable action, such as walking slowly or carrying equipment.

[0028] Alternatively, assume a firefighter is performing a firefighting task. Their gear is equipped with multiple flexible bending sensors, located on their left and right elbows, waist, and knees. These sensors are connected to the command center's monitoring system via Bluetooth. The sensors have a sampling rate of 50 Hz, collecting data 50 times per second. As the firefighter constantly moves and adjusts their posture during the firefighting process, the sensors collect their bending data in real time and transmit it to the monitoring system via Bluetooth. During this firefighting task, the monitoring system first filters the bending data transmitted by the sensors to remove any potential noise. For example, the bending angle data of the left elbow sensor are as follows: timestamp: 2025-06-28 14:30:00, bending angle: 30°; timestamp: 2025-06-28 14:30:01, bending angle: 45°; timestamp: 2025-06-28 14:30:02, bending angle: 60°; timestamp: 2025-06-28 14:30:03, bending angle: 75°; timestamp: 2025-06-28 14:30:04, bending angle: 90°; the monitoring system analyzes these data and finds that the bending angle of the left elbow increases from 30° to 90° in a short period of time, indicating that the firefighter may be performing an action that requires a large bending of the left arm, such as lifting a fire extinguisher. The processed data is stored in the database, and a bending angle matching table is collected. The bending angle matching table is shown in Table 1: Table 1 Bending angle matching table

[0029] The monitoring system's interface displays a real-time curve of the firefighter's left elbow's bending angle. The curve shows that between 2:30:00 PM and 2:30:04 PM on June 28, 2025, the elbow's bending angle rapidly increased from 30° to 90°, at a rate of 15° / s. Analyzing this data, the monitoring system determines that the firefighter may be performing an action requiring a rapid lift of his left arm, such as raising a fire extinguisher to put out a fire. The system also displays bending data from other sensors. For example, the waist sensor records an increase in bending angle from 15° to 30°, indicating the firefighter may be bending over. The knee sensors record bending angles of 90° and 100°, respectively, indicating the firefighter may be squatting. By comprehensively analyzing this data, the monitoring system can provide real-time insights into the firefighter's overall posture.

[0030] In step S12, when the firefighter performs a firefighting action, a bending change map of the flexible bending sensor is determined based on the change of the bending data of the flexible bending sensor; In the specific implementation process of the present invention, the specific steps are: S121: collecting firefighting training subjects of firefighters, determining firefighting training content of firefighters based on analysis of firefighting training subjects, matching corresponding firefighting actions based on firefighting training content, and then the firefighters perform the firefighting actions; S122: monitoring the firefighter's execution process of the firefighting action in real time, marking the corresponding action time period, and determining a plurality of bending data of the flexible bending sensor according to the database of the flexible bending sensor and the corresponding action time period, wherein the plurality of bending data changes along the time sequence of the action time period; S123: As the bending data of the flexible bending sensor changes, corresponding bending data tags are gradually constructed based on the respective bending data and corresponding time nodes, and a bending change graph of the flexible bending sensor is dynamically constructed according to the presentation of the multiple bending data tags in the same coordinate system.

[0031] In an embodiment of the present application, training subjects can be obtained through a fire training management system (such as an electronic training record system), or manually input by a coach or firefighter himself to enter the training subjects he is currently participating in. Training subjects are usually described in text form, such as "indoor fire fighting training", "high-rise building rescue training", "physical training", etc. The training management system parses the specific training content based on the name and description of the training subject. For example, "indoor fire fighting training" may include the following contents: using a fire extinguisher to extinguish indoor fires; checking the source of fire and assessing the fire; quickly evacuating the fire scene; and rescuing trapped people.

[0032] Establish an action library containing various firefighting actions. Each action has a detailed description and corresponding sensor data features. The action library may include the following: raising hands to operate a fire extinguisher; bending over to check the source of the fire; running quickly; climbing a ladder; carrying equipment; squatting to rescue trapped people. Based on the training content, match the specific actions that firefighters need to perform from the action library. For example, for "using a fire extinguisher to extinguish an indoor fire", the matching actions may be "raising hands to operate a fire extinguisher" and "bending over to check the source of the fire."

[0033] Firefighters are guided to perform matching firefighting actions through voice prompts, screen displays, or AR devices. For example, the system can give a voice prompt: "Please raise your hands to operate the fire extinguisher" and display the correct posture of the action. While the firefighters are performing the action, the system uses cameras and flexible bending sensors to monitor the execution of the action in real time to ensure that the firefighters' actions meet the training requirements.

[0034] Alternatively, suppose a firefighter is participating in "Indoor Firefighting Training." The training management system records the training subject as "Indoor Firefighting Training" and stores it in the database. The training management system analyzes the "Indoor Firefighting Training" subject and determines the training content includes: using a fire extinguisher to extinguish an indoor fire; inspecting the fire source and assessing the fire intensity; quickly evacuating the fire scene; and rescuing trapped individuals. Based on the training content "Using a fire extinguisher to extinguish an indoor fire," the system matches the following firefighting actions from the action library: Action 1: Raise your hand to operate the fire extinguisher (left elbow bent 90°, right elbow bent 90°); Action 2: Bend at the waist to inspect the fire source (waist bent 30°). The system instructs the firefighter to perform the "Raise your hand to operate the fire extinguisher" action through voice prompts. Following the prompts, the firefighter bends their left and right elbows to 90°, simulating the action of operating a fire extinguisher. Simultaneously, the system monitors the firefighter's execution of these actions in real time using cameras and flexible bending sensors to ensure compliance with requirements.

[0035] Furthermore, the execution process of the firefighting action by the firefighters is monitored in real time, and the corresponding action time period is marked. The multiple bending data of the flexible bending sensor are determined according to the database of the flexible bending sensor and the corresponding action time period. The multiple bending data change along the time sequence in the action time period, which is compatible with the overall consideration of the database of the flexible bending sensor and the corresponding action time period, thereby ensuring the accuracy of the multiple bending data of the flexible bending sensor.

[0036] In this case, a camera or other visual monitoring device is used to capture the firefighter's movements in real time. The camera can be positioned appropriately within the training area to ensure full coverage of the firefighter's full-body movements. Flexible bending sensors collect real-time bending data from various body parts. The sensors transmit this data to the monitoring system via wireless communication (e.g., Bluetooth). The start and end times of the firefighter's movements are determined using monitoring devices (e.g., cameras) or sensor data. Thresholds (e.g., changes in bending angle) can be set to automatically detect the start and end of movements. The start and end times of each movement are recorded in the monitoring system to form a time period. For example, movement 1 starts at 2:30:00 PM and ends at 2:30:05 PM.

[0037] Extract bending data from the flexible bending sensor database for the corresponding action time period. The database stores the bending angles of each sensor at different time points. This time series reflects the bending changes in various body parts during the firefighter's maneuvers. Arrange the extracted bending data in chronological order to form a time series. Time series analysis allows observation of the bending changes in various body parts during the firefighter's maneuvers. Visualize the time series data, for example, by drawing a broken line.

[0038] Specifically, suppose a firefighter is performing the "raise your hand to operate a fire extinguisher" maneuver. Multiple cameras are installed in the training area to monitor the firefighter's movements in real time. The firefighter is also fitted with multiple flexible bending sensors located on the left elbow, right elbow, waist, left knee, and right knee. These sensors collect bending data in real time at a 50Hz sampling rate and transmit the data to a monitoring system via Bluetooth. When the firefighter performs the "raise your hand to operate a fire extinguisher" maneuver, the monitoring system determines the start time of the maneuver as 2:30:00 PM and the end time as 2:30:05 PM based on the camera and sensor data. The monitoring system records this time period and marks it as the "raise your hand to operate a fire extinguisher" maneuver.

[0039] The monitoring system extracts bending data from the flexible bending sensor database for the period from 14:30:00 to 14:30:05. The extracted data is shown in Table 2: Table 2 Data Table

[0040]

[0041] It can be seen from the above table that when the firefighters performed the “raise hands to operate the fire extinguisher” action, the bending angles of the left and right elbows gradually increased from 0° to 90° and remained at 90° after 14:30:00.200, while the bending angles of the waist and knees remained basically unchanged.

[0042] Therefore, as the bending data of the flexible bending sensor changes, the corresponding bending data markers are gradually constructed based on the various bending data and the corresponding time nodes. The bending change graph of the flexible bending sensor is dynamically constructed according to the presentation of multiple bending data markers in the same coordinate system. The presentation of multiple bending data markers in the same coordinate system is introduced to dynamically construct the bending change graph of the flexible bending sensor.

[0043] A bending data tag refers to the sensor bending angle and position information recorded at a specific time. Each tag includes a timestamp, sensor location (e.g., left elbow, right elbow, waist), and corresponding bending angle. As sensor data is collected in real time, the system records the bending angle at each time point and generates a corresponding tag. For example, if the left elbow bending angle is 0° at the timestamp of 14:30:00.000, the tag (14:30:00.000, left elbow, 0°) is generated.

[0044] Define a coordinate system with time on the horizontal axis and bending angle on the vertical axis. Each sensor's data marker is represented in the coordinate system using a different color or line to distinguish it. Plot the bending data markers at each time point in the coordinate system to form a line graph or curve chart. The chart dynamically updates over time, displaying changes in bending angle in real time.

[0045] Alternatively, suppose firefighter Xiao Zhang is performing the "raise your hand to operate a fire extinguisher" action, and the flexible bending sensor collects data at a sampling rate of 50 Hz. A data-timestamp matching table is collected, as shown in Table 3: Table 3 Data-timestamp matching table

[0046] The corresponding marks are as follows: (14:30:00.000, left elbow, 0°); (14:30:00.020, left elbow, 10°); (14:30:00.040, left elbow, 20°); (14:30:00.060, left elbow, 30°); (14:30:00.080, left elbow, 40°); (14:30:00.100, left elbow, 50°); (14:30:00.120, left elbow, 60°); (14:30:00.140, left elbow, 70°); (14:30:00.160, left elbow, 80°); (14:30:00.180, left elbow, 90°).

[0047] In step S13, an abnormal posture node is determined based on the bending change graph, the action path of the firefighting action, and the current posture diagram of the firefighter; In the specific implementation process of the present invention, the specific steps are: S131: collecting the bending change graph, determining a plurality of abnormal bending data based on the identification of the bending change graph, and determining an abnormal bending area based on the values ​​and positions of the plurality of abnormal bending data and the corresponding body position of the firefighter; S132: collecting a current posture diagram of the firefighter, determining a corresponding matching area based on matching between the current posture diagram of the firefighter and the abnormal bending area, and marking a plurality of first abnormal posture sub-nodes based on autonomous identification of the matching area; S133: Monitor the firefighting action online, determine the action path of the firefighting action based on the online monitoring of the firefighting action, and determine multiple second sub-abnormal posture nodes according to the action path of the firefighting action and the firefighter's current posture diagram, and determine each abnormal posture node based on the matching of multiple first sub-abnormal posture nodes and multiple second sub-abnormal posture nodes.

[0048] In the embodiments of the present application, a bending change graph generated during a firefighter's maneuver is obtained from a monitoring system. This bending change graph records the changes in the bending angle of each sensor at different time points. The bending change graph is typically stored as a time series, recording the bending angle of each sensor at each time point. The data can be formatted as a table to facilitate subsequent analysis.

[0049] Abnormal bending data refers to bending angle data that does not conform to the normal motion path. Abnormal data can be identified by setting thresholds (e.g., bending angle changes that are too rapid or outside the normal range). Machine learning algorithms (such as anomaly detection algorithms) or simple threshold judgment methods are used to identify abnormal data. For example, if the waist bending angle increases from 0° to 30° in a short period of time, this can be considered abnormal data. An abnormal bending area refers to the area where the bending angle of a certain part of the firefighter's body changes abnormally during the execution of the action. This area can be determined by the value and location of the abnormal data. The abnormal bending area is determined based on the timestamp and sensor location of the abnormal data. For example, if the waist sensor records abnormal data, the abnormal bending area is the waist.

[0050] Alternatively, assuming that firefighter Xiao Zhang is performing the action of "raising his hand to operate the fire extinguisher", the monitoring system generates the following bending change graph, as shown in Table 4: Table 4 Bending change diagram

[0051]

[0052] Analysis of the aforementioned bending change graph revealed a sudden increase in the waist bending angle to 30° between 14:30:00.200 and 14:30:00.300, which is inconsistent with the normal motion path and therefore identified as abnormal bending data. The waist sensor recorded abnormal data (increase in waist bending angle from 0° to 30°), indicating that the abnormal bending area is the waist. Specifically, the left and right sides of the waist exhibit significant changes in the waist bending angle at these locations. This clearly demonstrates how to collect the bending change graph, identify abnormal bending data, and determine the abnormal bending area. This process provides an important foundation for subsequent identification of abnormal posture nodes.

[0053] Furthermore, a schematic diagram of the firefighter's current posture is collected, and the corresponding matching area is determined based on the matching of the firefighter's current posture schematic diagram and the abnormal bending area, and multiple first sub-abnormal posture nodes are marked based on the autonomous identification of the matching area, which is compatible with the overall consideration of the matching of the firefighter's current posture schematic diagram and the abnormal bending area, and ensures the accuracy of the corresponding matching area.

[0054] At this point, a diagram of the firefighter's current posture is obtained from the monitoring system. This diagram can be a two-dimensional image or a three-dimensional model, showing the position and posture of each body part. The diagram can be generated using image data captured by the camera and combined with a deep learning algorithm (such as OpenPose). The diagram typically contains the following information: the position coordinates of key body parts (such as the head, shoulders, elbows, waist, knees, ankles, etc.); the bending angles of each body part; and a timestamp for synchronization with the bending change diagram. The abnormal bending area (such as the waist) is matched with the current diagram to determine the corresponding matching area. The matching area refers to the body part corresponding to the abnormal bending area in the current diagram.

[0055] Use deep learning algorithms (such as convolutional neural networks (CNNs)) to autonomously identify the matching area and mark abnormal posture nodes. Abnormal posture nodes are specific locations within the matching area where posture is abnormal. The location and posture information of each abnormal posture node are marked. For example, the bending angle and position coordinates of the left and right sides of the waist are marked.

[0056] Alternatively, assume that firefighter Xiao Zhang is performing the action of "raising his hand to operate a fire extinguisher". The monitoring system collects his posture image through the camera and generates a posture diagram, which is shown in Table 5: Table 5 Posture diagram

[0057] The posture diagram shows that Xiao Zhang's left and right elbows are bent at 90°, his waist is bent at 30°, and his left and right knees are bent at 0°. In this example, the abnormal bending area is the waist (the waist bending angle increases from 0° to 30°). By matching the current posture diagram, the matching area is determined to be the waist, specifically the left and right sides of the waist. Within the matching area (waist), the autonomous recognition algorithm marks the following first-child abnormal posture nodes: Node 1: Left waist, position (180, 300), bending angle 30°; Node 2: Right waist, position (220, 300), bending angle 30°.

[0058] Therefore, the firefighting action is monitored online, and the action path of the firefighting action is determined based on the online monitoring of the firefighting action. According to the action path of the firefighting action and the firefighter's current posture schematic diagram, multiple second sub-abnormal posture nodes are determined. Based on the matching of multiple first sub-abnormal posture nodes and multiple second sub-abnormal posture nodes, each abnormal posture node is determined. The bending change graph of the flexible bending sensor is introduced, which is compatible with the overall consideration of the bending change graph, the action path of the firefighting action and the firefighter's current posture schematic diagram, thereby improving the monitoring accuracy of the abnormal posture nodes.

[0059] At this point, cameras or other visual monitoring equipment are used to monitor the firefighters' execution of firefighting maneuvers in real time. The cameras can be fixed at a suitable location within the training area to ensure full capture of the firefighters' full-body movements. Using image processing and posture recognition algorithms (such as OpenPose or YOLO), the motion trajectories of the firefighter's body parts are extracted in real time to form a motion path. The motion path includes the position coordinates of each key body part at different time points. The motion path is analyzed to identify parts that deviate from the normal motion path. For example, if the waist position changes abnormally during the motion path, this may indicate an abnormal waist posture. The abnormal part in the motion path is matched with the current posture diagram to determine the corresponding second-child abnormal posture node. The second-child abnormal posture node is the specific location in the motion path where the posture is abnormal. The first-child abnormal posture node and the second-child abnormal posture node are matched to determine the final abnormal posture node. The matching method can be based on position and posture information. For example, if two nodes have similar position and posture information, they are considered to be the same abnormal posture node.

[0060] Alternatively, suppose firefighter Xiao Zhang is performing the "raise your hand to operate a fire extinguisher" action. Multiple cameras are installed in the training area to monitor Xiao Zhang's actions in real time. The monitoring system uses the image data collected by the cameras and the OpenPose algorithm to generate a motion path matching table. The motion path matching table is shown in Table 6: Table 6 Action path matching table

[0061] The motion path shows an abnormal change in waist position between 14:30:00.200 and 14:30:00.300 (the waist bending angle increased from 0° to 30°). By matching the current posture diagram, the second abnormal posture node is identified as follows: Node 3: Left waist, position (180, 300), bending angle 30°; Node 4: Right waist, position (220, 300), bending angle 30°.

[0062] In the above example, the first sub-abnormal posture node and the second sub-abnormal posture node are as follows: first sub-abnormal posture node: Node 1: left side of waist, position (180, 300), bending angle 30°; Node 2: right side of waist, position (220, 300), bending angle 30°; second sub-abnormal posture node: Node 3: left side of waist, position (180, 300), bending angle 30°; Node 4: right side of waist, position (220, 300), bending angle 30°; through matching, the final abnormal posture nodes are determined as follows: Abnormal posture node 1: left side of waist, position (180, 300), bending angle 30°; Abnormal posture node 2: right side of waist, position (220, 300), bending angle 30°.

[0063] In step S14, during the monitoring process of the abnormal posture node, the corresponding abnormal motion path is determined based on the bending data and motion position of the abnormal posture node, and the abnormal area is determined based on the multiple abnormal motion paths, the corresponding motion environment and the current position of the firefighter; In the specific implementation process of the present invention, the specific steps are: S141: Real-time monitoring of abnormal posture nodes, collecting bending data of the abnormal posture nodes, marking the motion positions of the abnormal posture nodes, determining corresponding multimodal data based on the fusion of the bending data and the motion positions of the abnormal posture nodes, and determining the corresponding abnormal motion path based on the recognition of the multimodal data; S142: Performing a surrounding environment detection on the abnormal motion path, determining a plurality of environmental data along the detection of the surrounding environment of the abnormal motion path, determining a corresponding motion environment based on the plurality of environmental data and the abnormal motion path, and determining a first abnormal portion based on the plurality of abnormal motion paths and the corresponding motion environments; S143: Determine a second abnormal part according to the multiple abnormal action paths and the current position of the firefighter; and determine the abnormal area based on a mapping relationship among the first abnormal part, the second abnormal part, and the abnormal area.

[0064] In an embodiment of the present application, abnormal posture nodes are monitored in real time, and bending data of abnormal posture nodes are collected, and the action positions of abnormal posture nodes are marked. Corresponding multimodal data are determined based on the fusion of bending data and action positions of abnormal posture nodes. The corresponding abnormal action path is determined based on the identification of the multimodal data, which is compatible with the overall consideration of the fusion of bending data and action positions of abnormal posture nodes, thereby ensuring the accuracy of the corresponding multimodal data.

[0065] At this point, monitoring equipment such as flexible bending sensors and cameras are used to monitor the dynamic changes of abnormal posture nodes in real time. An abnormal posture node refers to a specific location where the posture is abnormal during the execution of an action. The flexible bending sensor collects bending angle data for the abnormal posture node. The camera collects movement position data for the abnormal posture node. The flexible bending sensor collects bending angle data of the abnormal posture node in real time. This data includes timestamp, sensor position, and corresponding bending angle. The collected bending data is recorded in the monitoring system's database.

[0066] Using image data captured by the camera, the specific locations of abnormal posture nodes during the motion process are marked. The motion location can be represented by the coordinates of key parts. The marked motion location data is recorded in the monitoring system's database. The bending data and motion location data are fused to form multimodal data. Multimodal data can include timestamps, sensor positions, bending angles, and motion location coordinates.

[0067] Alternatively, assume that when firefighter Xiao Zhang performs the "raise hands to operate a fire extinguisher" action, the monitoring system identifies abnormal posture nodes on the left and right sides of the waist. The monitoring system monitors these nodes in real time using flexible bending sensors and cameras. The bending data matching table collected by the monitoring system is shown in Table 7: Table 7 Bending data matching table

[0068] The action position data matching table marked by the monitoring system is shown in Table 8: Table 8 Action position data matching table

[0069] The fused multimodal data is shown in Table 9: Table 9 Action position data matching table

[0070] By analyzing the multimodal data, the abnormal motion paths were determined as follows: Abnormal motion path 1: the left side of the waist (node ​​1) was fixed at (180, 300) from 14:30:00.200 to 14:30:00.240, with a bending angle of 30°; Abnormal motion path 2: the right side of the waist (node ​​2) was fixed at (220, 300) from 14:30:00.200 to 14:30:00.240, with a bending angle of 30°.

[0071] Furthermore, the surrounding environment of the abnormal action path is detected, and multiple environmental data are determined along the detection of the surrounding environment of the abnormal action path. The corresponding action environment is determined based on the multiple environmental data and the abnormal action path. The first abnormal part is determined based on the multiple abnormal action paths and the corresponding action environment. The overall consideration of multiple environmental data and abnormal action paths is compatible to ensure the accuracy of the corresponding action environment.

[0072] At this point, by analyzing the multimodal data, the abnormal action path is determined as follows: Abnormal motion path 1: The left side of the waist (node ​​1) was fixed at (180, 300) from 14:30:00.200 to 14:30:00.240, with a bending angle of 30°.

[0073] Abnormal motion path 2: The right side of the waist (node ​​2) was fixed at (220, 300) from 14:30:00.200 to 14:30:00.240, with a bending angle of 30°.

[0074] At this point, environmental sensors (such as temperature sensors, smoke sensors, and cameras) are used to monitor the environment surrounding the abnormal motion path. These sensors can be fixed at appropriate locations within the training area or mounted on firefighters' equipment. Multiple environmental data points are collected along the abnormal motion path. This environmental data can include temperature, smoke concentration, and obstacle locations. The collected environmental data is organized into a table containing timestamps, sensor locations, temperature, smoke concentration, and obstacle locations. The environmental data at each time point is considered a data point.

[0075] The action environment refers to the environmental characteristics surrounding an abnormal action path during the execution of an action. This can include information such as temperature, smoke density, and obstacle locations. The environmental data is matched against the abnormal action path to determine the action environment corresponding to each abnormal action path. For example, if the environmental data surrounding an abnormal action path indicates high temperature, high smoke density, and the presence of an obstacle, the action environment for that path is "high temperature, high smoke density, and obstacle."

[0076] The first abnormal part refers to the abnormal area associated with the abnormal motion path within the motion environment. The first abnormal part can be a specific area or a set of features. The first abnormal part is determined by combining multiple abnormal motion paths and the corresponding motion environments. For example, if multiple abnormal motion paths all indicate a high temperature, high smoke concentration, and the presence of obstacles in a certain area, then this area can be considered the first abnormal part.

[0077] Alternatively, assume that when firefighter Xiao Zhang performs the "raise hands to operate a fire extinguisher" action, the monitoring system identifies abnormal movement paths on the left and right sides of the waist. The monitoring system uses environmental sensors to detect the environment around these paths. The data matching table collected by the environmental sensors is shown in Table 10: Table 10 Action position data matching table Data matching collected by environmental sensors

[0078] After sorting, the environmental data matching table is obtained, as shown in Table 11: Table 11 Environmental data matching table

[0079] Match the environmental data with the abnormal action paths to determine the action environment corresponding to each abnormal action path. For example, if the environmental data around an abnormal action path indicates high temperature, high smoke concentration, and the presence of obstacles, the action environment of this path is "high temperature, high smoke concentration, and obstacles." By analyzing the environmental data, the action environment is determined as follows: Action environment 1: Around the left side of the waist (node ​​1), the temperature is 50-54°C, the smoke density is 10-15%, and there is an obstacle at (160, 310). Action environment 2: Around the right side of the waist (node ​​2), the temperature is 50-54°C, the smoke density is 10-15%, and there is an obstacle at (240, 310). The first abnormal part refers to the abnormality in the action environment. By analyzing the environmental data, the action environment is determined as follows: Action environment 1: Around the left side of the waist (node ​​1), the temperature is 50-54°C, the smoke density is 10-15%, and there is an obstacle at (160, 310). Action environment 2: Around the right side of the waist (node ​​2), the temperature is 50-54°C, the smoke density is 10-15%, and there is an obstacle at (240, 310). It can be a feature set. Through comprehensive analysis of the abnormal movement path and movement environment, the first abnormal part is determined as follows: The first abnormal part: the area around the left and right sides of the waist, with a higher temperature (50-54°C), a higher smoke concentration (10-15%), and obstacles at positions (160, 310) and (240,310).

[0080] Therefore, the second abnormal part is determined according to multiple abnormal action paths and the current position of the firefighter; the abnormal area is determined based on the mapping relationship between the first abnormal part, the second abnormal part and the abnormal area, which is compatible with the overall consideration of the mapping relationship between the first abnormal part, the second abnormal part and the abnormal area, and ensures the accuracy of the abnormal area.

[0081] At this point, positioning technology (such as GPS, UWB, and Bluetooth beacons) is used to obtain the firefighter's current location in real time. This location information includes coordinates (x, y) and timestamps. Multiple abnormal motion paths are combined to analyze their relationship with the firefighter's current location. An abnormal motion path refers to an abnormal movement trajectory of a part of the firefighter's body during an action. The second abnormal part refers to an abnormal area near the firefighter's current location that is related to the abnormal motion path. By analyzing the abnormal motion path and the firefighter's current location, the specific location and range of the second abnormal part are determined.

[0082] The first abnormal portion refers to an abnormal area within the action environment associated with an abnormal action path. Step S142 has already determined the specific location and characteristics of the first abnormal portion. A mapping relationship is established between the first abnormal portion and the second abnormal portion to determine the final abnormal area. The abnormal area refers to the area defined by the abnormal action path of a part of the firefighter's body and the characteristics of the surrounding environment during the action. The final abnormal area is determined by combining the first and second abnormal portions. The abnormal area can be a specific area or a set of features.

[0083] Alternatively, assume the first abnormal portion is: First abnormal portion: The area around the left and right sides of the waist, with a higher temperature (50-54°C), a higher smoke density (10-15%), and the presence of obstacles at (160, 310) and (240, 310). The second abnormal portion is: Second abnormal portion: The area around the firefighter's current position (200, 300), specifically the area around the left and right sides of the waist. By combining the first and second abnormal portions, the final abnormal region is determined as follows: Assume the first abnormal portion is: First abnormal portion: The area around the left and right sides of the waist, with a higher temperature (50-54°C), a higher smoke density (10-15%), and the presence of obstacles at (160, 310) and (240, 310). The second abnormal portion is: Second abnormal portion: The area around the firefighter's current position (200, 300), specifically the area around the left and right sides of the waist. By combining the first abnormal part and the second abnormal part, the final abnormal area is determined as follows: Abnormal area: the area around the left and right sides of the waist, and the area around the firefighter's current position (200, 300), with a higher temperature (50-54°C), a higher smoke concentration (10-15%), and the presence of obstacles at positions (160, 310) and (240, 310).

[0084] In step S15, the firefighter's risky action is determined based on the abnormal area and the corresponding firefighting action, and the firefighter's posture correction measures are determined based on the risky action, the corresponding posture data, and the bending data of the flexible bending sensor, and the firefighter's dynamic adjustment of the risky action is triggered; In the specific implementation process of the present invention, the specific steps are: S151: Acquire abnormal areas, locate corresponding firefighting actions based on the abnormal areas, action paths, and action times of the firefighting actions, determine multiple action nodes based on the detection of the firefighting actions, determine risk coefficients based on the positions, action ranges, and action environments of the action nodes, determine risk nodes for firefighters based on the comparison of the risk coefficients with corresponding risk coefficient thresholds, and output the risk actions corresponding to the risk nodes; S152: Determine corresponding posture data based on posture detection of the risk node, determine a first correction range based on the risk action and the corresponding posture data, and determine a second correction range based on the risk action and bending data of the flexible bending sensor; S153: Determine the firefighter's posture correction measures based on the mapping relationship between the first correction range, the second correction range, and the posture correction measures, and mark the corrective action steps corresponding to the posture correction measures to trigger the firefighter's dynamic adjustment of the risky action.

[0085] In an embodiment of the present application, abnormal areas are collected, and corresponding firefighting actions are located based on the training of abnormal areas, action paths and action times of firefighting actions. Multiple action nodes are determined based on the detection of firefighting actions, and the risk coefficient is determined based on the position, action range and action environment of the action node. The risk node of the firefighter is determined based on the comparison between the risk coefficient and the corresponding risk coefficient threshold, and the risk action corresponding to the risk node is output, which is compatible with the overall consideration of the comparison between the risk coefficient and the corresponding risk coefficient threshold, thereby ensuring the accuracy of the risk node of the firefighter.

[0086] In this case, an abnormal area is defined by the abnormal movement path of a firefighter's body part during the execution of the action and the surrounding environmental characteristics. Characteristics of the abnormal area include location, temperature, smoke density, and obstacle location. Information about the abnormal area, including location, temperature, and smoke density, is obtained from the monitoring system. Environmental sensors (such as temperature sensors, smoke sensors, and cameras) are used to collect this data in real time.

[0087] The action path refers to the trajectory of various body parts during a firefighter's action. The action time refers to the time period during which the action is performed. Using historical data and training models, we analyze the relationship between abnormal areas, action paths, and action time. We then use machine learning algorithms (such as decision trees and support vector machines) to locate the corresponding firefighting action.

[0088] Action nodes refer to the position and posture information of key parts during an action. Action nodes can be detected using camera and sensor data. Using cameras and flexible bending sensors, the position and posture of key parts of firefighters are monitored in real time. The timestamp, position coordinates, and bending angle of each action node are recorded.

[0089] The risk factor is a quantitative indicator used to assess the risk level of an action node. The risk factor is calculated using a weighted score, taking into account factors such as location, range of action, and action environment. Each factor is assigned a weight, adjusting its contribution to the overall score based on its importance. The risk factor is calculated for each action node. The risk factor threshold is a preset value used to determine whether an action node is a risky node. If the risk factor exceeds the threshold, the node is considered a risky node. The risk factor of each action node is compared to the risk factor threshold to identify the risky node.

[0090] Alternatively, assume that the monitoring system identifies the following abnormal areas: Abnormal Area: The area around the left and right sides of the waist, with high temperature (50-54°C), high smoke concentration (10-15%), and obstacles at (160, 310) and (240, 310). Assume that, based on the training data, the monitoring system identifies the corresponding firefighting action as "Raise Hand to Operate Fire Extinguisher." The action path and timing are as follows: Action Path: The left and right sides of the waist are located at (180, 300) and (220, 300), respectively, with a bending angle of 30°. Action Time: 2:30:00.200 to 2:30:00.240. The monitoring system detects the following action nodes: Action Node 1: Left side of the waist, position (180, 300), bending angle 30°; Action Node 2: Right side of the waist, position (220, 300), bending angle 30°.

[0091] Assume the following weight distribution: position weight: 0.4; motion range weight: 0.3; motion environment weight: 0.3; the risk factor of action node 1 is calculated as follows: position score: (180, 300) is close to the obstacle (160, 310), score 80; motion range score: bending angle of 30° exceeds the normal range, score 70; motion environment score: temperature 52°C, smoke concentration 12%, score 75; risk factor = (80 × 0.4) + (70 × 0.3) + (75 × 0.3) = 32 + 21 + 22.5 = 75.5 Similarly, the risk factor for action node 2 is 74.5. Assume the risk factor threshold is 70. The risk factor for action node 1 is 75.5, exceeding the threshold of 70 and thus identified as a risk node. The risk factor for action node 2 is 74.5, exceeding the threshold of 70 and thus identified as a risk node. The firefighting action corresponding to risk nodes 1 and 2 is "excessive bending of the waist," which may result in a waist injury. The monitoring system outputs the following risk action: Risk action: Excessive bending of the waist, which may result in a waist injury.

[0092] Furthermore, the corresponding posture data is determined based on the posture detection of the risk node, the first correction range is determined according to the risk action and the corresponding posture data, and the second correction range is determined according to the risk action and the bending data of the flexible bending sensor, which is compatible with the overall consideration of the risk action and the bending data of the flexible bending sensor to ensure the accuracy of the second correction range.

[0093] At this point, posture detection involves real-time monitoring of the position and posture of various parts of a firefighter's body through camera and sensor data. Posture data includes the position coordinates and bending angles of key parts. Using cameras and flexible bending sensors, posture data of risk nodes is collected in real time. The timestamp, position coordinates, and bending angle of each risk node are recorded. The first correction range refers to the position range of key parts in the posture data that require correction. This range is determined based on the risky action and posture data to ensure that the firefighter's movements meet safety standards. The risky action and posture data are analyzed to determine the position range of key parts that require correction.

[0094] For example, if the waist bend angle is too large, it needs to be adjusted to a safe range.

[0095] Alternatively, assuming the risky action is "excessive waist bending," the corresponding posture data is as follows: Risk Node 1: Left side of waist, position (180, 300), bending angle 30°; Risk Node 2: Right side of waist, position (220, 300), bending angle 30°. Based on the risky action and posture data, the first correction range is determined as follows: First Correction Range: Left and right waist position range (170, 290) to (230, 310). Assuming the risky action is "excessive waist bending," the corresponding bending data is as follows: Risk Node 1: Left side of waist, bending angle 30°; Risk Node 2: Right side of waist, bending angle 30°. Based on the risky action and bending data, the second correction range is determined as follows: Second Correction Range: Left and right waist bending angle range 0° to 15°.

[0096] A flexible bending sensor is used to collect the bending angles of key areas in real time. For example, the bending angles on the left side of the waist are 30°, and on the right side of the waist are 30°. Risk Node 1: Left side of the waist, position (180, 300), bending angle 30°; Risk Node 2: Right side of the waist, position (220, 300), bending angle 30°; First Correction Range: Left and right sides of the waist position range (170, 290) to (230, 310); Risk Node 1: Left side of the waist, bending angle 30°; Risk Node 2: Right side of the waist, bending angle 30°; Second Correction Range: Left and right sides of the waist bending angle range 0° to 15°.

[0097] Therefore, based on the mapping relationship between the first correction range, the second correction range and the posture correction measures, the firefighter's posture correction measures are determined, and the corrective action steps corresponding to the posture correction measures are marked to trigger the firefighter's dynamic adjustment of the risky action. This is compatible with the overall consideration of the first correction range, the second correction range and the mapping relationship between the posture correction measures, ensuring the accuracy of the firefighter's posture correction measures. At the same time, it realizes further control of the risky action, and then realizes the overall consideration of the risky action, the corresponding posture data and the bending data of the flexible bending sensor, ensuring the accuracy of the firefighter's posture correction measures, and realizing the dynamic adjustment of the risky action by the firefighter.

[0098] In this case, the posture correction measure mapping relationship refers to mapping the first correction range and the second correction range to specific corrective measures. This mapping relationship can be implemented using preset rules or a machine learning model. Based on the first correction range (position range) and the second correction range (bending angle range), combined with the posture correction measure mapping relationship, a specific corrective measure is determined. The corrective measure should include specific movement adjustment suggestions, such as adjusting the bending angle or position.

[0099] Corrective action steps refer to the specific steps that need to be followed when executing corrective measures. Each corrective measure can be broken down into multiple specific action steps. Label the specific steps of each corrective action according to the corrective measure. The corrective action step should describe in detail how to adjust the posture, including the starting position, target position, and adjustment method of the action. Dynamic adjustment refers to adjusting the firefighter's movements based on real-time data during the execution of the action. Dynamic adjustment can remind firefighters to adjust their posture in real time through the monitoring system. The monitoring system monitors the firefighter's posture data in real time. When a risky action is detected, the corrective action is triggered and a real-time reminder is sent to the firefighter.

[0100] Specifically, assume that the first correction range and the second correction range are as follows: first correction range: the position range of the left and right sides of the waist is (170, 290) to (230, 310); second correction range: the bending angle range of the left and right sides of the waist is 0° to 15°; the posture correction measure mapping relationship is as follows: Correction measure 1: left side of the waist, the bending angle is adjusted to 0° to 15°; Correction measure 2: right side of the waist, the bending angle is adjusted to 0° to 15°; the determined posture correction measures are as follows: Correction measure 1: left side of the waist, the bending angle is adjusted to 0° to 15°; Correction measure 2: right side of the waist, the bending angle is adjusted to 0° to 15°.

[0101] Assume that the corrective measures are as follows: Corrective measure 1: Left side of the waist, adjust the bending angle to 0° to 15°; Corrective measure 2: Right side of the waist, adjust the bending angle to 0° to 15°; the marked corrective action steps are as follows: Corrective action step 1: Left side of the waist, slowly adjust the bending angle to 0° to 15°; Starting position: 30°; Target position: 0° to 15°, Adjustment method: Slowly relax the waist and avoid sudden movements.

[0102] Corrective Action Step 2: Slowly adjust the right side of the waist to a bend angle between 0° and 15°; Starting position: 30°; Target position: 0° to 15°; Adjustment method: Slowly relax the waist and avoid sudden movements. Suppose the monitoring system detects that firefighter Xiao Zhang's left and right waist bend angles are 30°, respectively, exceeding the second correction range (0° to 15°). The monitoring system triggers corrective action and sends a real-time alert to the firefighter: Real-time alert: The left and right waist bend angles are too large. Please slowly adjust to 0° to 15° to avoid waist injury. The detailed analysis and practical examples above clearly demonstrate how the mapping between the first correction range, the second correction range, and the posture corrective action determines the firefighter's posture corrective action and marks the corrective action step, triggering the firefighter to dynamically adjust to the risky action.

[0103] See also Figure 2 , Figure 2 : is a schematic diagram of the structure of a firefighter posture monitoring system based on a flexible bending sensor in an embodiment of the present invention; the firefighter posture monitoring system based on a flexible bending sensor includes: a posture diagram module 21 for determining a firefighter's current posture diagram based on the firefighter's current posture data and body shape data, and marking bending data of a flexible bending sensor disposed on the firefighter's clothes; a bending change map module 22 for determining a bending change map of the flexible bending sensor based on changes in bending data of the flexible bending sensor when a firefighter performs a firefighting action; An abnormal posture node module 23 is used to determine an abnormal posture node according to the bending change diagram, the action path of the firefighting action and the current posture diagram of the firefighter; The abnormal area module 24 is used to determine the corresponding abnormal motion path based on the bending data and motion position of the abnormal posture node during the monitoring process of the abnormal posture node, and determine the abnormal area according to multiple abnormal motion paths, corresponding motion environments and the current position of the firefighter; The posture corrective measures module 25 is used to determine the firefighter's risky actions based on the abnormal area and the corresponding firefighting actions, and determine the firefighter's posture corrective measures based on the risky actions, the corresponding posture data and the bending data of the flexible bending sensor, and trigger the firefighter's dynamic adjustment of the risky actions.

[0104] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A firefighter posture monitoring method based on a flexible bending sensor, characterized in that: include: Determine a current posture diagram of the firefighter based on the current posture data and body shape data of the firefighter, and mark bending data of a flexible bending sensor configured on the firefighter's clothes; determining a bending change map of the flexible bending sensor based on changes in bending data of the flexible bending sensor when a firefighter performs a firefighting action; determining abnormal posture nodes according to the bending change graph, the action path of the firefighting action, and the current posture diagram of the firefighter; During the monitoring process of the abnormal posture node, the corresponding abnormal motion path is determined based on the bending data and motion position of the abnormal posture node, and the abnormal area is determined based on the multiple abnormal motion paths, the corresponding motion environment and the current position of the firefighter; The firefighter's risky action is determined based on the abnormal area and the corresponding firefighting action, and the firefighter's posture corrective measures are determined based on the risky action, the corresponding posture data and the bending data of the flexible bending sensor, and the firefighter's dynamic adjustment of the risky action is triggered.

2. The firefighter posture monitoring method based on the flexible bending sensor according to claim 1 is characterized in that: The method of determining the firefighter's current posture diagram based on the firefighter's current posture data and body shape data, and marking the bending data of the flexible bending sensor configured on the firefighter's clothes, includes: The firefighter's current position is collected, and the corresponding target detection area is determined based on the firefighter's current position and surrounding cameras. The firefighter's current posture data is determined based on real-time detection of the target detection area. A schematic diagram of the firefighter's current posture is constructed based on the current posture data, the corresponding time node, and the firefighter's body contour image. Determining key posture parts of the firefighter based on the detection of the firefighter's current posture diagram, and determining the flexible bending sensor configured on the firefighter's clothes based on the position detection of the key posture parts; The flexible bending sensor is monitored in real time; and bending data recorded by the flexible bending sensor as the firefighter's posture changes are determined based on the real-time monitoring of the flexible bending sensor, wherein the bending data shows the degree of bending of the flexible bending sensor and the degree of change in the firefighter's posture.

3. The firefighter posture monitoring method based on the flexible bending sensor according to claim 1 is characterized in that: The method of determining a bending change graph of the flexible bending sensor based on a change in bending data of the flexible bending sensor when a firefighter performs a firefighting action includes: Collecting firefighters' fire training subjects, and determining firefighters' fire training content based on the analysis of the firefighters' fire training subjects, matching corresponding firefighting actions based on the firefighters' fire training content, and then the firefighters perform the firefighting actions; The execution process of the firefighting action by the firefighter is monitored in real time, and the corresponding action time period is marked. A plurality of bending data of the flexible bending sensor is determined according to the database of the flexible bending sensor and the corresponding action time period, and the plurality of bending data changes along the time sequence of the action time period; As the bending data of the flexible bending sensor changes, corresponding bending data tags are gradually constructed based on each bending data and the corresponding time node, and the bending change graph of the flexible bending sensor is dynamically constructed according to the presentation of multiple bending data tags in the same coordinate system.

4. The firefighter posture monitoring method based on flexible bending sensor according to claim 1 is characterized in that: The determining of the abnormal posture node according to the bending change graph, the action path of the firefighting action, and the firefighter's current posture schematic diagram includes: The bending change map is collected, a plurality of abnormal bending data are determined based on the identification of the bending change map, and an abnormal bending area is determined based on the values ​​and positions of the plurality of abnormal bending data and the corresponding body positions of the firefighter.

5. The firefighter posture monitoring method based on flexible bending sensor according to claim 4 is characterized in that: The method of determining the abnormal posture node according to the bending change graph, the action path of the firefighting action, and the firefighter's current posture diagram further includes: Collecting a firefighter's current posture schematic diagram, determining a corresponding matching area based on a match between the firefighter's current posture schematic diagram and the abnormal bending area, and marking a plurality of first sub-abnormal posture nodes based on autonomous identification of the matching area; The firefighting action is monitored online, and the action path of the firefighting action is determined based on the online monitoring of the firefighting action. A plurality of second sub-abnormal posture nodes are determined according to the action path of the firefighting action and the current posture diagram of the firefighter, and each abnormal posture node is determined based on the matching of the plurality of first sub-abnormal posture nodes and the plurality of second sub-abnormal posture nodes.

6. The firefighter posture monitoring method based on flexible bending sensor according to claim 1 is characterized in that: During the monitoring process of the abnormal posture node, the corresponding abnormal motion path is determined based on the bending data and motion position of the abnormal posture node, and the abnormal area is determined according to the multiple abnormal motion paths, the corresponding motion environment and the current position of the firefighter, including: Abnormal posture nodes are monitored in real time, and the bending data of abnormal posture nodes are collected, and the action positions of abnormal posture nodes are marked. The corresponding multimodal data is determined based on the fusion of the bending data and action positions of the abnormal posture nodes, and the corresponding abnormal action path is determined based on the recognition of the multimodal data.

7. The firefighter posture monitoring method based on the flexible bending sensor according to claim 6 is characterized in that: In the process of monitoring the abnormal posture node, the corresponding abnormal motion path is determined based on the bending data and motion position of the abnormal posture node, and the abnormal area is determined according to the multiple abnormal motion paths, the corresponding motion environment and the current position of the firefighter, further comprising: Performing a surrounding environment detection on the abnormal motion path, determining a plurality of environmental data along the detection of the surrounding environment of the abnormal motion path, determining a corresponding motion environment based on the plurality of environmental data and the abnormal motion path, and determining a first abnormal portion based on the plurality of abnormal motion paths and the corresponding motion environments; The second abnormal part is determined according to the multiple abnormal action paths and the current position of the firefighter; and the abnormal area is determined based on the mapping relationship among the first abnormal part, the second abnormal part and the abnormal area.

8. The firefighter posture monitoring method based on flexible bending sensor according to claim 1 is characterized in that: The method of determining a firefighter's risky action based on the abnormal area and the corresponding firefighting action, determining the firefighter's posture corrective measures based on the risky action, the corresponding posture data, and the bending data of the flexible bending sensor, and triggering the firefighter to dynamically adjust the risky action includes: Abnormal areas are collected, and corresponding firefighting actions are located based on the training of abnormal areas, action paths and action times of firefighting actions. Multiple action nodes are determined based on the detection of firefighting actions, and the risk coefficient is determined based on the position, action range and action environment of the action node. The risk node of the firefighter is determined based on the comparison of the risk coefficient with the corresponding risk coefficient threshold, and the risk action corresponding to the risk node is output.

9. The firefighter posture monitoring method based on flexible bending sensor according to claim 8, characterized in that: The method further includes determining a risky action of a firefighter based on the abnormal area and the corresponding firefighting action, determining a posture corrective measure of the firefighter based on the risky action, the corresponding posture data, and the bending data of the flexible bending sensor, and triggering the firefighter to dynamically adjust the risky action. Determining corresponding posture data based on posture detection of the risk node, determining a first correction range based on the risk action and the corresponding posture data, and determining a second correction range based on the risk action and bending data of the flexible bending sensor; The firefighter's posture corrective measures are determined based on the mapping relationship among the first correction range, the second correction range and the posture corrective measures, and the corrective action steps corresponding to the posture corrective measures are marked to trigger the firefighter's dynamic adjustment of the risky action.

10. A firefighter posture monitoring system based on a flexible bending sensor, characterized in that: The firefighter posture monitoring system based on the flexible bending sensor is applied to the firefighter posture monitoring method based on the flexible bending sensor as described in any one of claims 1 to 9. The firefighter posture monitoring system based on the flexible bending sensor includes: a posture diagram module, configured to determine a current posture diagram of the firefighter based on the current posture data and body shape data of the firefighter, and mark bending data of a flexible bending sensor disposed on the firefighter's clothes; a bending change map module for determining a bending change map of the flexible bending sensor based on changes in bending data of the flexible bending sensor when a firefighter performs a firefighting action; An abnormal posture node module is used to determine an abnormal posture node according to the bending change graph, the action path of the firefighting action and the current posture diagram of the firefighter; An abnormal area module is used to determine the corresponding abnormal motion path based on the bending data and motion position of the abnormal posture node during the monitoring process of the abnormal posture node, and determine the abnormal area according to multiple abnormal motion paths, corresponding motion environments and the current position of the firefighter; The posture corrective measures module is used to determine the firefighter's risky actions based on the abnormal area and the corresponding firefighting actions, and to determine the firefighter's posture corrective measures based on the risky actions, the corresponding posture data and the bending data of the flexible bending sensor, and to trigger the firefighter's dynamic adjustment of the risky actions.

Images