Riding helmet motion signal steering control method, system and equipment and medium

Abstract

The invention relates to the technical field of intelligent wearable equipment, in particular to a riding helmet motion signal steering control method, system and equipment and a medium. The method comprises the following steps: acquiring head movement data and environment state data of a rider; inputting the head motion data and the environment state data into a preset first motion recognition model; and outputting a control parameter for controlling the display state of the turn signal lamp according to the first action recognition model. The head movement data and the environment state data comprise head posture data collected by a multi-axis accelerometer and a gyroscope, light intensity data collected by an environment light sensor, front road condition image data collected by a micro camera and real-time riding speed data. Through multi-source data fusion and intelligent algorithm processing, accurate steering intention recognition and self-adaptive display control are realized, and riding safety is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent wearable devices, and in particular to a method, system, device and medium for controlling the steering of motion signals of a cycling helmet. Background Art

[0002] With the rapid development of urban intelligent transportation and the popularization of bicycle travel, cycling safety issues have become increasingly prominent. Traditional cycling turn prompts mainly rely on hand gestures. This method not only requires the rider to let go of the handlebars to signal when turning, which increases the risk of riding, but also has poor visibility in low light or bad weather conditions. In recent years, smart helmet technology has made significant progress, and various helmet products that integrate LED (light-emitting diode) signal lights, communication modules and sensors have been launched one after another. However, most of the smart helmets on the market currently use buttons or voice control to achieve the turn prompt function. This interactive method has a slow response speed in emergency situations and is easily disturbed by environmental noise, which cannot meet the rider's dual needs for safety and convenience. In addition, existing smart helmets generally lack the ability to perceive the riding environment and cannot adaptively adjust the display parameters of the signal lights according to factors such as ambient light intensity and road conditions. This greatly limits its practicality in complex and changeable actual riding scenarios.

[0003] In response to the above problems, the industry has already had many improvement plans, such as head posture detection based on inertial measurement unit (IMU), action recognition based on deep learning, and other technologies gradually applied to the field of smart helmets. However, existing head action recognition algorithms often only focus on a single motion feature and fail to fully utilize multi-source sensor data for collaborative perception, resulting in insufficient recognition accuracy. Summary of the invention

[0004] Based on this, it is necessary to propose a cycling helmet motion signal steering control method, system, device and medium to address the problems of inaccurate turn signal light control and poor safety performance in the prior art.

[0005] A cycling helmet motion signal steering control method, the method comprising:

[0006] Obtain the rider's head movement data and environmental status data;

[0007] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0008] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0009] Preferably, the head movement data includes head posture data, the environmental status data includes light intensity data, front road condition image data and real-time riding speed data, and the step of obtaining the rider's head movement data and environmental status data includes:

[0010] Collecting the head posture data by means of a multi-axis accelerometer and a gyroscope;

[0011] Collecting the light intensity data through an ambient light sensor;

[0012] Collecting the front road condition image data by using a micro camera;

[0013] Collecting the real-time riding speed data through a speed sensor;

[0014] The head movement data and environmental status data are collected in real time at a preset sampling frequency.

[0015] Preferably, the step of inputting the head movement data and the environmental state data into a preset first action recognition model comprises:

[0016] Determining whether the head posture data reaches a preset turning trigger threshold;

[0017] When the head posture data reaches the preset turning trigger threshold, acquiring the ambient light intensity data and the front road condition image data;

[0018] The data combination consisting of the head posture data, riding speed data, light intensity data and front road condition image data is input into the first action recognition model.

[0019] Preferably, the control parameter for controlling the display state of the turn signal lamp according to the output of the first action recognition model includes:

[0020] Controlling the turning direction of the turn signal lamp using a turning control parameter corresponding to the head posture data and output by the first action recognition model;

[0021] Controlling the brightness of the turn signal lamp using a brightness control parameter corresponding to the ambient light intensity data and output by the first action recognition model;

[0022] The flashing frequency of the turn signal light is controlled by a frequency control parameter corresponding to the riding speed data and output by the first action recognition model.

[0023] Preferably, the method further comprises:

[0024] Determining whether there is an obstacle in front of the rider through the front road condition image data;

[0025] When the obstacle is detected, a vibration or sound prompt is triggered, and the triggering of the turn signal light is delayed;

[0026] When it is detected that the obstacle does not exist, it is re-determined whether a preset turning trigger threshold is reached.

[0027] Preferably, the method further comprises:

[0028] Establishing a wireless connection with a mobile terminal;

[0029] Receiving preset parameters set by a user through the mobile terminal, the preset parameters including a preset turn trigger threshold;

[0030] Sending real-time status information to the mobile terminal, the status information including the helmet power level, current steering status and warning information.

[0031] Preferably, the first action recognition model includes an input layer, an intermediate layer and an output layer, and the control parameters for controlling the display state of the turn signal lamp according to the output of the first action recognition model include:

[0032] Receiving head posture data, ambient light intensity data, front road condition image data and riding speed data through the input layer;

[0033] Using a neural network through the intermediate layer to perform action intention recognition;

[0034] A steering control signal is generated through the output layer, and the control signal includes a steering direction, a signal light brightness, and a flashing frequency.

[0035] On the other hand, a cycling helmet motion signal steering control system is provided. Based on the cycling helmet motion signal steering control method as described above, the cycling helmet motion signal steering control system comprises:

[0036] Data collection module, to obtain rider's head movement data and environmental status data;

[0037] An action recognition module inputs the head movement data and the environmental state data into a preset first action recognition model;

[0038] A signal control module is used to control the control parameters of the display state of the turn signal light according to the output of the first action recognition model.

[0039] Another aspect provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps:

[0040] Obtain the rider's head movement data and environmental status data;

[0041] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0042] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0043] Another aspect provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor performs the following steps:

[0044] Obtain the rider's head movement data and environmental status data;

[0045] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0046] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0047] The beneficial effects of the present invention are as follows: the present invention realizes multi-dimensional perception of riding scenes by acquiring head motion data and environmental status data, avoiding the limitation of traditional solutions that rely on only a single sensor, thereby improving the accuracy of turn signal control. In general, the present invention provides an innovative smart helmet interaction solution through multi-source data fusion, intelligent algorithm processing and human-computer interaction optimization, opening up a new direction for the application of smart wearable devices in sports scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 A schematic diagram of a flow chart of a cycling helmet motion signal steering control method provided in an embodiment of the present application;

[0050] Figure 2 Another schematic diagram of a flow chart of a method for controlling the steering of a motion signal of a cycling helmet provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of the structure of a cycling helmet motion signal steering control system provided in an embodiment of the present application.

[0052] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Reference Figure 1-2 , which is the first embodiment of the present invention, provides a cycling helmet motion signal steering control method, comprising:

[0055] S100: Acquiring rider's head movement data and environmental status data;

[0056] Specifically, the head motion data is the head posture data collected by the multi-axis accelerometer and gyroscope, including the pitch angle, yaw angle and roll angle data of the head, and the environmental status data includes the ambient light intensity data, the front road condition image data and the real-time riding speed data. The ambient light intensity data is collected by the ambient light sensor with a range of 0-65535l ux, the front road condition image data is collected by the micro camera with a collection frame rate of 30fps and an image resolution of 1920×1080, and the riding speed data is collected in real time by the speed sensor with a sampling frequency of 10Hz.

[0057] In an optional embodiment, the head posture data, the ambient light intensity data, the front road condition image data and the real-time riding speed data can be any combination of data acquired by different sensors integrated in the helmet. For example, when the head posture data is collected by a nine-axis IMU sensor, the riding speed data can be acquired by a GPS (Global Positioning System, abbreviated as GPS, global positioning system) module.

[0058] In an optional embodiment, head posture data, ambient light intensity data, front road condition image data, and real-time riding speed data can also add or reduce other sensor data according to different riding scenarios. For example, when more refined safety protection is needed, heart rate data (i.e., the fifth category of physiological data) can be added to monitor the rider's physical condition, thereby further improving safety and comprehensiveness. In addition, considering the impact of different weather conditions on sensor data collection, an adaptive data filtering algorithm can also be introduced to dynamically calibrate the collected sensor data to ensure that high-quality status data can be obtained in various riding environments.

[0059] In an optional embodiment, if the riding speed does not need to be considered, the real-time riding speed data can be omitted, thereby reducing the workload of data collection and processing. However, in order to ensure the comprehensiveness of security in this application, such data is still retained.

[0060] It should be noted that the acquisition of the above sensor data can fully cover all key information during riding, including but not limited to head movement status, ambient light conditions, road conditions ahead, and riding speed, so as to ensure the accuracy and safety of steering control. At the same time, by introducing a variety of sensors and acquisition devices, such as IMU sensors, ambient light sensors, cameras, and speed sensors, the advantages of various types of data can be fully utilized, complement each other, and improve the accuracy and reliability of control. In addition, according to the needs of different riding scenarios, the types of data and sampling parameters can be flexibly adjusted to meet the needs of actual applications. This multi-source data fusion strategy not only improves the adaptability and robustness of the system, but also provides a rich data source for subsequent intelligent analysis.

[0061] In some embodiments, the head movement data and environmental status data include:

[0062] Collecting the head posture data by means of a multi-axis accelerometer and a gyroscope;

[0063] Collecting the light intensity data through an ambient light sensor;

[0064] Collecting the front road condition image data by using a micro camera;

[0065] Collecting the real-time riding speed data through a speed sensor;

[0066] The head movement data and environmental status data are collected in real time at a preset sampling frequency.

[0067] Specifically, the sampling frequency of head posture data is 100Hz to ensure that subtle changes in head movements can be captured in real time; the sampling frequency of ambient light intensity data is 1Hz to meet the response requirements for lighting changes; the sampling frequency of the road condition image data ahead is 30fps to ensure the continuity and smoothness of the image; the sampling frequency of real-time riding speed data is 10Hz, which can accurately reflect speed changes.

[0068] In an optional embodiment, the sampling frequencies of different sensors can be dynamically adjusted according to actual application requirements. For example, in complex road conditions, the sampling frame rate of the road image ahead can be increased to 60fps to obtain more detailed road condition information; when riding at night, the sampling frequency of the ambient light intensity sensor can be reduced to 0.2Hz to save system resources.

[0069] It should be noted that the setting of sampling frequency needs to strike a balance between data accuracy and system resource consumption. A sampling frequency that is too high will increase the data processing burden, while a sampling frequency that is too low may miss key information. This system has selected the optimal sampling frequency combination through experimental verification to ensure that the system can accurately capture the required information while maintaining low resource usage.

[0070] S200: inputting the head movement data and the environmental state data into a preset first action recognition model;

[0071] Specifically, the first action recognition model adopts a deep learning architecture based on LSTM (Long Short-Term Memory Network), which includes a temporal feature extraction module and an action classification module. The temporal feature extraction module consists of two layers of LSTM units, each containing 128 neurons, which are used to capture the temporal features of head movements; the action classification module contains two fully connected layers, with 64 and 32 neurons respectively, and finally outputs the action category probability through Softmax (normalized exponential function).

[0072] In an optional embodiment, the network structure of the first action recognition model can be adjusted according to actual needs. For example, an attention mechanism can be added to enhance the recognition ability of key action features, or a CNN-LSTM (hybrid neural network model) hybrid architecture can be used to better process image sequence data.

[0073] It should be noted that the design of the model needs to take into account real-time requirements. Although a complex network structure may provide higher recognition accuracy, it will also increase computing latency. The LSTM architecture used in this system achieves a good balance between accuracy and response speed.

[0074] In some embodiments, inputting the head movement data and the environmental state data into a preset first action recognition model includes:

[0075] Determine whether the head posture data reaches the preset turning trigger threshold;

[0076] When the preset turning trigger threshold is reached, the ambient light intensity data and the image data of the road condition ahead are obtained;

[0077] The acquired data combination is input into the first action recognition model.

[0078] Specifically, the system sets three key attitude angle thresholds: yaw angle threshold ±30° (turn left / right), pitch angle threshold ±20° (head up / head down), and roll angle threshold ±15° (head tilt). When any angle exceeds the corresponding threshold and lasts for more than 0.3 seconds, the turning intention judgment is triggered. At the same time, the system also sets an angular velocity threshold of 30° / s to filter out unintentional head movements caused by factors such as road bumps.

[0079] In an optional embodiment, the trigger threshold can be adjusted according to the user's habits. For example, for users who are accustomed to making small steering movements, the yaw angle threshold can be adjusted to ±25°; for users with a more stable riding posture, the angular velocity threshold can be reduced to 20° / s.

[0080] It should be noted that the setting of the threshold requires a balance between sensitivity and reliability. A threshold that is too low can easily lead to false triggering, while a threshold that is too high can reduce the responsiveness of the system. This system has determined the optimal threshold combination through a large amount of user test data, which can meet the needs of most users.

[0081] Determine whether the head posture data reaches the preset turning trigger threshold;

[0082] Specifically, the system sets three key attitude angle thresholds: yaw angle threshold ±30° (turn left / right), pitch angle threshold ±20° (head up / head down), and roll angle threshold ±15° (head tilt). When any angle exceeds the corresponding threshold and lasts for more than 0.3 seconds, the turning intention judgment is triggered. At the same time, the system also sets an angular velocity threshold of 30° / s to filter out unintentional head movements caused by factors such as road bumps.

[0083] In an optional embodiment, the trigger threshold can be adjusted according to the user's habits. For example, for users who are accustomed to making small steering movements, the yaw angle threshold can be adjusted to ±25°; for users with a more stable riding posture, the angular velocity threshold can be reduced to 20° / s.

[0084] It should be noted that the setting of the threshold requires a balance between sensitivity and reliability. A threshold that is too low can easily lead to false triggering, while a threshold that is too high can reduce the responsiveness of the system. This system has determined the optimal threshold combination through a large amount of user test data, which can meet the needs of most users.

[0085] In some embodiments, the first action recognition model includes:

[0086] The input layer receives head posture data, ambient light intensity data, front road condition image data and riding speed data;

[0087] The middle layer uses a neural network to recognize action intentions;

[0088] The output layer generates a steering control signal, which includes the steering direction, signal light brightness and flashing frequency.

[0089] When the preset turning trigger threshold is reached, the ambient light intensity data and the image data of the road condition ahead are obtained;

[0090] Specifically, when a turn intention is detected, the system immediately increases the sampling frequency of the ambient light sensor to 10Hz and continues to collect data for 2 seconds to accurately assess the current lighting conditions. At the same time, the high-speed image acquisition mode is activated to continuously capture 20 frames of high-definition road condition images (about 0.67 seconds) with a resolution of 2560×1440. The system also saves historical data 1 second before the trigger for contextual analysis.

[0091] In an optional embodiment, the data collection strategy can be adjusted according to different time periods. For example, when riding at night, the collection time of ambient light intensity data can be extended to 3 seconds to obtain a more stable lighting assessment; in conditions of low visibility such as rainy days, the frame rate of image collection can be increased to 60fps.

[0092] It should be noted that the timeliness and integrity of data collection are crucial for safety assessment. The system adopts a dynamic cache mechanism to ensure the real-time nature of data and avoid data loss.

[0093] Inputting the acquired data combination into a first action recognition model;

[0094] Specifically, the system first organizes the head posture data into a time series feature matrix (T×6 dimensions, T is the time window length, 6 is the three angles and their angular velocity), and the ambient light intensity data and riding speed data constitute the environmental feature vector (2×1 dimensions). The front road image data is extracted through the pre-trained ResNet-18 network to extract spatial features (512-dimensional vector). These features are combined into a unified input tensor through the feature fusion layer and sent to the LSTM network for time series analysis.

[0095] In an optional embodiment, the feature combination method can be optimized according to the actual effect. For example, an attention mechanism can be introduced to dynamically weight different features, or a multimodal feature fusion network can be used to improve the expressiveness of features.

[0096] It should be noted that the way of combining features directly affects the performance of the model. The feature fusion method adopted by this system not only retains the key information of various types of data, but also realizes efficient feature expression, providing a good feature foundation for subsequent action recognition.

[0097] S300: Control parameters for controlling the display state of the turn signal lamp according to the output of the first action recognition model.

[0098] Specifically, the model output includes the category probability of the turning intention (turn left / turn right / go straight) and the confidence score. When the probability of a turning intention exceeds 0.85 and the confidence score is greater than 0.9, the system activates the corresponding turn signal light. The control parameters (brightness, frequency) of the signal light are dynamically adjusted according to the ambient light intensity and riding speed, such as increasing the LED brightness in strong light and increasing the flashing frequency when riding at high speed.

[0099] Specifically, the system has designed an intelligent dimming mechanism to automatically adjust the brightness of the LED signal light according to the ambient light intensity data. When there is sufficient daylight (light intensity > 10000l ux), the LED works in the maximum power mode (300mcd); in normal daylight (1000-10000l ux), the LED uses medium power (200mcd); in twilight or indoor environments (<1000l ux), the LED uses lower power (100mcd) to avoid glare. The flicker frequency is positively correlated with the riding speed, with a base frequency of 1Hz, and a 0.2Hz frequency increase for every 5km / h increase in speed, with a maximum of no more than 2Hz.

[0100] In an optional embodiment, the LED display parameters can be adjusted according to the requirements of different scenarios. For example, in special environments such as tunnels, a high contrast mode can be enabled to increase the LED brightness while reducing the flashing frequency; in congested roads, the flashing frequency can be increased to improve the warning effect.

[0101] It should be noted that the display effect of the signal light needs to strike a balance between visibility and energy saving. This system has optimized various parameters through field testing to ensure that the ideal warning effect can be achieved under various lighting conditions.

[0102] In some embodiments, the control parameters for controlling the display state of the turn signal light according to the output of the first action recognition model include:

[0103] Controlling the turning direction of the turn signal lamp using a turning control parameter corresponding to the head posture data and output by the first action recognition model;

[0104] Controlling the brightness of the turn signal lamp using a brightness control parameter corresponding to the ambient light intensity data and output by the first action recognition model;

[0105] The flashing frequency of the turn signal light is controlled by a frequency control parameter corresponding to the riding speed data and output by the first action recognition model.

[0106] Specifically, the system determines the turning intention by analyzing the changing trend of the head posture data. For the intention to turn left, the characteristic is that the yaw angle deflects to the left by more than 30° and lasts for more than 300ms, and the roll angle changes by no more than ±5°; the intention to turn right is manifested as the yaw angle deflects to the right by more than 30° and lasts for more than 300ms, and the roll angle also remains within the range of ±5°. The system also introduces motion trajectory smoothness evaluation, which filters jitter and unintentional movements by calculating the angular velocity variance (<10° / s2).

[0107] In an optional embodiment, the steering judgment logic can be customized according to user habits. For example, a "fast mode" can be provided for experienced users, reducing the duration threshold to 200ms; and a "stable mode" can be provided for novice users, adding additional posture stability checks.

[0108] It should be noted that the accuracy of steering judgment is directly related to riding safety. The system adopts a multiple verification mechanism to ensure the accuracy of judgment and avoid safety hazards caused by false triggering.

[0109] In some embodiments, the method further comprises:

[0110] Determine whether there is an obstacle through the image data of the road ahead;

[0111] When an obstacle is detected, a vibration or sound prompt is triggered and the triggering of the turn signal light is delayed;

[0112] When it is detected that the obstacle does not exist, it is re-determined whether a preset turning trigger threshold is reached.

[0113] Adjust the brightness of turn signal lights based on ambient light intensity data;

[0114] Specifically, the system uses a piecewise function to adjust the LED brightness:

[0115] Strong light environment (20000l ux): LED working current 350mA, brightness 300mcd;

[0116] Normal daylight (5000-20000l ux): LED working current 250mA, brightness 200mcd;

[0117] Cloudy day / dusk (500-5000l ux): LED working current 150mA, brightness 100mcd;

[0118] Night / Indoor (<500l ux): LED working current 80mA, brightness 50mcd.

[0119] At the same time, the system implements a gradual dimming function, with a switching time of 200ms between adjacent brightness levels to avoid visual discomfort caused by sudden changes.

[0120] In an optional embodiment, the brightness adjustment strategy can be optimized in combination with weather information. For example, in rainy and foggy weather, the reference brightness of each gear can be appropriately increased; on urban roads at night, the brightness can be dynamically adjusted according to street lighting conditions.

[0121] It should be noted that LED brightness adjustment must ensure the visibility of the signal while avoiding interference to other road users. This system has been tested extensively in the field to optimize the brightness parameters under various lighting conditions.

[0122] In some embodiments, the method further comprises:

[0123] Establishing a wireless connection with a mobile terminal;

[0124] Receiving preset parameters set by a user through a mobile terminal, the preset parameters including a preset turning trigger threshold;

[0125] Send real-time status information to the mobile terminal, including the helmet battery level, current steering status and warning information.

[0126] Adjust the flashing frequency of the turn signal lights according to riding speed data;

[0127] Specifically, the system designs a speed-adaptive flicker control strategy:

[0128] Low speed section (0-10km / h): basic frequency 0.8Hz;

[0129] Medium speed section (10-20km / h): linearly increase to 1.2Hz;

[0130] High speed section (20-30km / h): linearly increases to 1.6Hz;

[0131] Ultra-high speed section (30km / h): fixed at 2.0Hz.

[0132] The flickering waveform uses a modified square wave with a lighting time ratio (duty cycle) of 60%, which can reduce visual fatigue while ensuring recognition.

[0133] In an optional embodiment, the flashing parameters can be adjusted according to the riding scene. For example, on a bicycle lane, a lower basic frequency can be used; on a road with mixed traffic of motor vehicles, the flashing frequency can be increased to enhance the warning effect.

[0134] It should be noted that the setting of the flicker frequency needs to take into account the visual characteristics and reaction time of the human eye. Flickering too fast may cause visual fatigue, while flickering too slow may cause the key prompts to be missed. This system has determined the optimal frequency parameters through human factors engineering research.

[0135] In some embodiments, judging whether there is an obstacle based on the front road condition image data specifically includes:

[0136] The system uses a target detection algorithm based on YOLOv5 to perform real-time analysis of the road conditions ahead. The detection targets include:

[0137] Static obstacles: roadblocks, construction facilities, parked vehicles, etc.;

[0138] Dynamic obstacles: pedestrians, bicycles, motor vehicles, etc.;

[0139] Road conditions: water accumulation, damaged road surface, speed bumps, etc.

[0140] The detection result includes the target category, location coordinates and confidence level. When an obstacle is detected and the confidence level exceeds 0.85, the system enters the warning state.

[0141] In an optional embodiment, the obstacle detection strategy can be optimized according to the riding environment. For example, the detection confidence threshold can be lowered to 0.75 at night to increase the safety margin; the detection range can be expanded in crowded areas to detect potential risks in advance.

[0142] It should be noted that the real-time and accuracy of obstacle detection is crucial to avoiding collisions. The system achieves millisecond-level detection response through edge computing technology, providing a guarantee for timely avoidance of danger.

[0143] The entire embodiment realizes a safe and reliable riding steering prompt function through multi-source data fusion and intelligent algorithm processing, providing riders with an intuitive and intelligent auxiliary tool. The various modules of the system work closely together, ensuring both functional integrity and ease of use.

[0144] Furthermore, this embodiment also provides a cycling helmet motion signal steering control system, comprising:

[0145] Data collection module, to obtain rider's head movement data and environmental status data;

[0146] The motion recognition module inputs the head motion data and the environmental state data into a preset first motion recognition model;

[0147] The signal control module is used to control the control parameters of the display state of the turn signal lamp according to the output of the first action recognition model.

[0148] In summary, the beneficial effects of collaborative acquisition and processing of multi-source data are:

[0149] By simultaneously acquiring head posture data, ambient light intensity data, front road image data, and real-time riding speed data, multi-dimensional perception of riding scenes is achieved. This multi-source data collaborative collection and processing mechanism avoids the limitations of traditional solutions that rely on only a single sensor, greatly improves the environmental adaptability and reliability of the system, and significantly improves the accuracy of turning intention recognition.

[0150] Beneficial effects of adaptive sampling strategy:

[0151] By setting different sampling frequencies for different types of data (head posture, ambient light intensity, image, speed) and dynamically adjusting according to scene requirements, on-demand allocation of data collection is achieved. This adaptive sampling mechanism not only ensures the real-time nature of key data, but also avoids resource waste, effectively reduces system power consumption, and prolongs the use time of the helmet.

[0152] Beneficial effects of LSTM-based action recognition model:

[0153] The double-layer LSTM network structure, combined with the temporal feature extraction and action classification modules, can effectively capture the temporal features of head movements. This deep learning architecture not only improves the accuracy of steering intention recognition, but also effectively filters unintentional movements, significantly reducing the false trigger rate.

[0154] Beneficial effects of multiple threshold joint judgment:

[0155] By setting the angle thresholds of yaw angle, pitch angle and roll angle, and combining them with the angular velocity threshold for comprehensive judgment, a complete steering intention judgment mechanism is formed. This multi-threshold joint judgment strategy effectively solves the problem of false triggering caused by factors such as road bumps in traditional solutions, and improves the reliability of the system under complex road conditions.

[0156] Beneficial effects of intelligent display control:

[0157] Based on the ambient light intensity and riding speed, the LED signal light brightness and flashing frequency are dynamically adjusted. This adaptive display control not only ensures the visibility of the signal, but also avoids interference with other road users. Especially when riding at night, through precise brightness control, the warning effect is improved while effectively reducing light pollution.

[0158] Beneficial effects of active safety warning:

[0159] By analyzing the road conditions ahead in real time and combining it with the target detection algorithm based on YOLOv5, it can identify potential dangers in advance. This active safety warning mechanism effectively shortens the reaction time for danger avoidance and significantly improves riding safety.

[0160] Beneficial effects of human-computer interaction optimization:

[0161] The mobile terminal APP enables personalized parameter settings and status monitoring, providing customized experience for different users. This flexible interaction method not only improves the user experience, but also achieves continuous optimization of the system through data collection and analysis, greatly improving user satisfaction.

[0162] In general, the present invention successfully solves the safety hazards and inconveniences of traditional riding steering prompts through multi-source data fusion, intelligent algorithm processing and human-computer interaction optimization, and achieves a comprehensive improvement in safety, reliability and user experience. This solution not only provides a new interactive paradigm for smart helmets, but also opens up a new direction for the application of smart wearable devices in sports scenes.

[0163] This embodiment also provides a computer device, which is applicable to a method for controlling the steering of a motion signal of a cycling helmet, and includes a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement a method for controlling the steering of a motion signal of a cycling helmet as proposed in the above embodiment:

[0164] Obtain the rider's head movement data and environmental status data;

[0165] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0166] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0167] The present invention achieves multi-dimensional perception of riding scenes by acquiring head motion data and environmental status data, avoiding the limitation of traditional solutions that rely on only a single sensor, thereby improving the accuracy of turn signal control. In general, the present invention provides an innovative smart helmet interaction solution through multi-source data fusion, intelligent algorithm processing and human-computer interaction optimization, opening up a new direction for the application of smart wearable devices in sports scenes.

[0168] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, a method for controlling the steering of a motion signal of a cycling helmet as proposed in the above embodiment is implemented:

[0169] Obtain the rider's head movement data and environmental status data;

[0170] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0171] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0172] The present invention achieves multi-dimensional perception of riding scenes by acquiring head motion data and environmental status data, avoiding the limitation of traditional solutions that rely on only a single sensor, thereby improving the accuracy of turn signal control. In general, the present invention provides an innovative smart helmet interaction solution through multi-source data fusion, intelligent algorithm processing and human-computer interaction optimization, opening up a new direction for the application of smart wearable devices in sports scenes.

[0173] like Figure 4 As shown, this embodiment also provides a computer device, which can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless method can be implemented through WI FI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball or a touch pad set on the housing of the computer device, or an external keyboard, touch pad or mouse, etc.

[0174] The computer device can be used to implement a cycling helmet motion signal steering control method proposed in the above embodiment:

[0175] Obtain the rider's head movement data and environmental status data;

[0176] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0177] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0178] The present invention achieves multi-dimensional perception of riding scenes by acquiring head motion data and environmental status data, avoiding the limitation of traditional solutions that rely on only a single sensor, thereby improving the accuracy of turn signal control. In general, the present invention provides an innovative smart helmet interaction solution through multi-source data fusion, intelligent algorithm processing and human-computer interaction optimization, opening up a new direction for the application of smart wearable devices in sports scenes.

[0179] This embodiment also provides a computer-readable storage medium. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, etc. Various media that can store program codes. More specific examples (non-exhaustive list) of computer-readable storage media include the following: an electrical connection portion (electronic device) with one or more wirings, a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and editable read-only memory (EPROM or flash memory), an optical fiber device, and a portable optical disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0180] The storage medium stores a computer program, which, when executed by the processor, implements a method for controlling the steering of a motion signal of a cycling helmet as proposed in the above embodiment:

[0181] Obtain the rider's head movement data and environmental status data;

[0182] Inputting the head movement data and the environmental state data into a preset first action recognition model;

[0183] A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

[0184] The present invention achieves multi-dimensional perception of riding scenes by acquiring head motion data and environmental status data, avoiding the limitation of traditional solutions that rely on only a single sensor, thereby improving the accuracy of turn signal control. In general, the present invention provides an innovative smart helmet interaction solution through multi-source data fusion, intelligent algorithm processing and human-computer interaction optimization, opening up a new direction for the application of smart wearable devices in sports scenes.

[0185] It should be understood that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0186] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0187] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0188] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0189] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible 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.

[0190] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A cycling helmet motion signal steering control method, characterized in that: include: Obtain the rider's head movement data and environmental status data; Inputting the head movement data and the environmental state data into a preset first action recognition model; A control parameter for controlling a display state of a turn signal lamp according to an output of the first action recognition model.

2. A cycling helmet motion signal steering control method as claimed in claim 1, characterized in that: The head movement data includes head posture data, the environmental status data includes light intensity data, front road condition image data and real-time riding speed data, and the step of obtaining the rider's head movement data and environmental status data includes: Collecting the head posture data by means of a multi-axis accelerometer and a gyroscope; Collecting the light intensity data through an ambient light sensor; Collecting the front road condition image data by using a micro camera; Collecting the real-time riding speed data through a speed sensor; The head movement data and environmental status data are collected in real time at a preset sampling frequency.

3. A cycling helmet motion signal steering control method as claimed in claim 2, characterized in that: The step of inputting the head movement data and the environmental state data into a preset first action recognition model comprises: Determining whether the head posture data reaches a preset turning trigger threshold; When the head posture data reaches the preset turning trigger threshold, acquiring the ambient light intensity data and the front road condition image data; The data combination consisting of the head posture data, riding speed data, light intensity data and front road condition image data is input into the first action recognition model.

4. A cycling helmet motion signal steering control method as claimed in claim 3, characterized in that: The control parameter for controlling the display state of the turn signal lamp according to the output of the first action recognition model includes: Controlling the turning direction of the turn signal lamp using a turning control parameter corresponding to the head posture data and output by the first action recognition model; Controlling the brightness of the turn signal lamp using a brightness control parameter corresponding to the ambient light intensity data and output by the first action recognition model; The flashing frequency of the turn signal light is controlled by a frequency control parameter corresponding to the riding speed data and output by the first action recognition model.

5. A cycling helmet motion signal steering control method as claimed in claim 4, characterized in that: The method further comprises: Determining whether there is an obstacle in front of the rider through the front road condition image data; When the obstacle is detected, a vibration or sound prompt is triggered, and the triggering of the turn signal light is delayed; When it is detected that the obstacle does not exist, it is re-determined whether a preset turning trigger threshold is reached.

6. A cycling helmet motion signal steering control method as claimed in claim 5, characterized in that: The method further comprises: Establishing a wireless connection with a mobile terminal; receiving a restriction parameter set by the mobile terminal, the restriction parameter including a preset turn triggering threshold; Sending real-time status information to the mobile terminal, the status information including the helmet power level, current steering status and warning information.

7. A cycling helmet motion signal steering control method as claimed in claim 6, characterized in that: The first action recognition model includes an input layer, an intermediate layer and an output layer. The control parameters for controlling the display state of the turn signal lamp according to the output of the first action recognition model include: Receiving head posture data, ambient light intensity data, front road condition image data and riding speed data through the input layer; Using a neural network through the intermediate layer to perform action intention recognition; A steering control signal is generated through the output layer, and the control signal includes a steering direction, a signal light brightness, and a flashing frequency.

8. A cycling helmet motion signal steering control system, based on the cycling helmet motion signal steering control method according to any one of claims 1 to 7, characterized in that: The riding helmet motion signal steering control system comprises: Data collection module, to obtain rider's head movement data and environmental status data; An action recognition module inputs the head movement data and the environmental state data into a preset first action recognition model; A signal control module is used to control the control parameters of the display state of the turn signal light according to the output of the first action recognition model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the cycling helmet motion signal steering control method described in any one of claims 1-7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cycling helmet motion signal steering control method described in any one of claims 1 to 7 are implemented.

Images