Safety warning device, method, electronic device, and computer readable medium

Abstract

The application provides a safety warning device, method, electronic equipment and computer readable medium, belongs to the field of intelligent driving, collects environment data, the environment data includes: distance, speed, visual image and three-dimensional terrain; receiving the environment data, carrying out fusion analysis, risk assessment and decision-making on the environment data; according to the decision instruction, starting from information warning to progressive response of active control, the safety warning device provided by the application fuses distance, speed, visual image and three-dimensional terrain and other data by multidimensional data collection, overcomes the single sensor bottleneck of tunnel environment light change, weather influence and the like, and realizes reliable monitoring under 24 hours, multiple weather conditions.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to a safety warning device, method, electronic device, and computer-readable medium. Background Technology

[0002] With the development of intelligent connected vehicle technology, vehicle-road-cloud integrated systems have become a core technological path to improve traffic safety (refer to the "Guidelines for the Construction and Application of Vehicle-Road-Cloud Integrated Systems (Official Release) - October 2024"). Tunnels, ramps, and other special scenarios, due to their complex environments (such as sudden changes in lighting, slope variations, and limited visibility), have become high-risk areas for traffic accidents. Existing technologies have the following problems in these scenarios:

[0003] 1. Inappropriate sensor deployment

[0004] Traditional solutions often use a single camera or radar, without optimization for specific scenarios. For example, relying solely on cameras in tunnels is susceptible to backlighting and dust, resulting in target recognition accuracy below 70%; ramps lack dedicated slope sensors, relying solely on vehicle speed to assess the risk of slippage, leading to a false alarm rate as high as 30%.

[0005] 2. Poor adaptability of the recognition algorithm

[0006] Existing algorithms fail to distinguish scene features: the traditional frame difference method is used to identify abnormal parking in tunnels, with a false negative rate of over 25%; the judgment of vehicle slippage on slopes does not integrate the features of slope and sudden speed changes, resulting in a warning delay of ≥500ms on steep slope sections.

[0007] 3. Fixed warning threshold

[0008] Traditional systems use static thresholds (such as triggering warnings at a fixed vehicle speed > 60 km / h) without considering dynamic factors such as traffic density and weather. When the traffic flow in tunnels exceeds 80 vehicles / hour, the false alarm rate rises to 5 times / day; in rainy or foggy weather, the missed detection rate increases by 40% due to the failure to adjust the threshold (refer to "Insufficient Adaptation to Dynamic Scenarios" in Part 1 of "Vehicle-Road-Cloud Integrated System: System Composition and Basic Platform Architecture").

[0009] 4. Insufficient multi-sensor collaboration

[0010] In the existing technology, radar and camera data are not synchronized in time (error > 50ms), resulting in a deviation of > 2m in the fusion result, which cannot meet the high-precision early warning requirements of tunnels / ramp (refer to the "Data Synchronization Requirements" in CSAE 295.3-2023 "Vehicle-Road-Cloud Integrated System Part 3: Road-Cloud Data Interaction Specification"). Summary of the Invention

[0011] The present invention aims to solve at least one of the technical problems existing in the prior art, and proposes a safety early warning device and method.

[0012] In a first aspect, embodiments of the present invention provide a safety warning device, comprising:

[0013] A sensor-coordinated sensing unit is used to collect environmental data, including distance, speed, visual images, and three-dimensional terrain.

[0014] An edge computing unit is used to receive the environmental data, perform fusion analysis, risk assessment, and make decisions based on the environmental data;

[0015] The multi-level early warning execution unit is used to initiate a progressive response from information early warning to active control based on decision instructions.

[0016] In some embodiments, the sensor collaborative sensing unit includes: millimeter-wave radar, infrared camera, and lidar;

[0017] Furthermore, the aforementioned lidar is used to generate a high-precision three-dimensional point cloud map by emitting a laser beam and receiving reflections, and to automatically switch to slope monitoring mode when a rapid change in slope is detected by sensing the slope and its rate of change.

[0018] In some embodiments, when the ambient temperature is >60°C, the lubricating oil of the lidar motor is perfluoropolyether; when the ambient temperature is <-20°C, the lidar point cloud data is processed by bilateral filtering and a point cloud interpolation algorithm is used.

[0019] In some embodiments, the edge computing unit includes:

[0020] Multi-source data fusion module: fuses LiDAR data and millimeter-wave radar data, and uses infrared images to perform target detection through the YOLOv8 model;

[0021] Risk assessment module: It integrates slope data and vehicle speed information from the vehicle CAN bus to calculate the speed change rate Δv and trajectory offset, and adopts a dual-threshold graded intervention.

[0022] Furthermore, when Δv > 15 km / h / s, a primary warning is issued, and the braking system is pre-charged.

[0023] When Δv > 20 km / h / s and the deviation > 1.5 meters, a high-level warning is issued, directly triggering ESP emergency braking;

[0024] Adaptive threshold adjustment module: Based on a fuzzy logic controller, it dynamically adjusts the radar detection threshold and camera confidence threshold according to real-time traffic density, ambient light, and weather conditions.

[0025] In some embodiments, the infrared image is used for target detection via a YOLOv8 model, including:

[0026] The backbone network of the YOLOv8 model introduces the GhostConv module, replacing the first convolutional layer;

[0027] Furthermore, in tunnel scenarios, when a stationary vehicle is detected to have exceeded a set time and is not located in a parking space, it is determined to be an abnormal parking situation.

[0028] In some embodiments, the multi-level early warning execution unit includes:

[0029] Yellow alert: Triggers in-vehicle audio and visual alerts to warn the driver;

[0030] Orange alert: Triggers in-vehicle audio and visual alerts to warn the driver, broadcasts lane-level speed limit information, and pre-charges the brakes on at-risk vehicles;

[0031] Red alert: Directly activates ESP to perform emergency braking, simultaneously turns on hazard warning lights, and synchronizes the alert to the cloud platform.

[0032] In some embodiments, the sensor-coordinated sensing unit includes:

[0033] Hardware timestamps are embedded during sensor data acquisition;

[0034] The ECU performs local clock calibration via PTP;

[0035] A dynamic time warping algorithm is used to align the time axis of multi-sensor data streams, with the error controlled within 10ms.

[0036] Thirdly, embodiments of the present invention also provide a security early warning method for implementing the above-described apparatus, comprising:

[0037] Collect environmental data, including: distance, speed, visual images, and three-dimensional terrain;

[0038] Receive the environmental data, perform fusion analysis and risk assessment on the environmental data, and make decisions;

[0039] Based on the decision-making instructions, a progressive response is initiated, from information early warning to proactive control.

[0040] Thirdly, embodiments of the present invention also provide an electronic device, comprising:

[0041] One or more processors;

[0042] Memory, used to store one or more programs;

[0043] When the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0044] Fourthly, embodiments of the present invention also provide a computer-readable medium storing a computer program, which, when executed by a processor, implements the steps in the method.

[0045] The safety early warning device provided by this invention collects environmental data, including distance, speed, visual images, and three-dimensional terrain. It receives the environmental data, performs fusion analysis, risk assessment, and makes decisions based on the data. According to the decision instructions, it initiates a progressive response from information warning to active control. This safety early warning device overcomes the bottleneck of single sensors due to changes in tunnel lighting and weather conditions by fusing multi-dimensional data collection, such as distance, speed, visual images, and three-dimensional terrain, achieving reliable monitoring 24 hours a day under various weather conditions. It completes complex data fusion and AI analysis locally with extremely low response latency, meeting the stringent real-time requirements of safety systems. Through fuzzy logic control, system parameters can be intelligently adjusted according to traffic and environmental conditions, ensuring high sensitivity while keeping the daily false alarm rate extremely low. It establishes a complete intervention chain from remote warning and near-end alert to direct vehicle control, realizing smooth and safe takeover under the "human-machine co-driving" concept, effectively preventing and mitigating traffic accidents in tunnel and ramp scenarios. It is suitable for safety monitoring and risk warning of intelligent connected vehicles in complex road conditions. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the structure of one embodiment of the safety warning device of the present invention;

[0047] Figure 2 This is an architectural diagram of one embodiment of the safety warning device of the present invention;

[0048] Figure 3 This is a flowchart of the Kalman filter optimization process of this invention;

[0049] Figure 4 This is a structural diagram of the fuzzy logic controller of the present invention;

[0050] Figure 5 This is a timeline diagram for the risk assessment of this invention;

[0051] Figure 6 This is a flowchart illustrating the safety early warning method of the present invention;

[0052] Figure 7 This is a schematic diagram of the structure of an embodiment of the electronic device of the present invention. Detailed Implementation

[0053] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0054] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0055] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0056] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0057] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0058] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0059] In related technologies, there are several issues: Single-sensor reliance: One brand of tunnel warning system uses only LiDAR, resulting in a false alarm rate as high as 12 times / day (in rainy weather) and a response delay of 200 ms. In low-light environments, the point cloud density of the LiDAR decreases by 40%, leading to an increased false alarm rate. Static threshold setting: Another slope monitoring system uses a slope > 8° to trigger warnings, resulting in a false alarm rate of 40% on continuous curves with slopes of 6-8°. The combined effect of vehicle speed and slope is not considered, leading to inaccurate risk assessment. Lack of scene adaptation: Traditional systems cannot adjust detection strategies based on traffic density, resulting in a 30% increase in false alarm rate during peak hours (vehicles > 100 vehicles / hour), and lacking a weather compensation mechanism.

[0060] The technical terms involved in this invention are as follows:

[0061] C-V2X Cellular-Vehicle to Everything ECU Edge Computing Unit OS-CFAR Ordered Statistics - Constant False Alarm Rate Algorithm YOLOv8 Deep learning-based object detection model (You Only Look Once v8) ESP Electronic Stability Program PTP Precision Time Protocol DTW Dynamic Time Warping DSRC Dedicated Short Range Communications Special scenarios This refers to traffic scenarios with complex environments and high accident rates, such as tunnels and ramps, characterized by sudden changes in lighting, gradient variations, and limited visibility. abnormal parking The vehicle remains stationary in a non-parking space area for more than 30 seconds. Risk of car rolling away A dangerous situation in a slope scenario where a vehicle's speed suddenly changes by Δv > 20 km / h / s and its trajectory deviates by > 1.5m. Dynamic threshold adjustment A mechanism that adjusts the warning triggering conditions in real time based on factors such as traffic density and weather. Multi-sensor fusion Spatiotemporal alignment and feature fusion of data from multiple devices such as millimeter-wave radar, infrared cameras, and lidar

[0062] To address at least one of the technical problems existing in the aforementioned related technologies, the present invention provides a safety early warning device. Figure 1 This is a schematic diagram of a safety early warning device provided in an embodiment of the present invention.

[0063] like Figure 1 and Figure 2 As shown, the safety early warning device includes: a sensor collaborative sensing unit, an edge computing unit, and a multi-level early warning execution unit. The specific implementation schemes of each unit are described in detail below.

[0064] The sensor collaborative sensing unit is used to collect environmental data, including distance, speed, visual images, and three-dimensional terrain.

[0065] In this embodiment, the specific components and implementation scheme of the sensor collaborative sensing unit are as follows:

[0066] Millimeter-wave radar: Responsible for detecting the distance, speed, and azimuth of moving and stationary targets at medium to long ranges. Employing advanced signal processing algorithms, it effectively filters out environmental clutter interference, reporting data only when the target's reflected signal is sufficiently strong, ensuring reliable detection. It achieves stable speed and distance measurement in adverse weather conditions such as rain, fog, and dust, providing the system with fundamental moving target data.

[0067] Infrared camera: Capable of imaging in extremely low light conditions, automatically activating infrared illumination in dimly lit tunnels or at night to acquire clear video images. Through intelligent analysis algorithms, it identifies vehicle outlines, parking status, and whether the parking space is compliant. This solves the problem of traditional cameras failing in insufficient lighting within tunnels, enabling 24 / 7 uninterrupted visual monitoring and accurate target classification (e.g., distinguishing between normal driving and abnormal parking).

[0068] LiDAR: Generates high-precision 3D point cloud maps by emitting laser beams and receiving reflections. Specifically designed to sense the slope and rate of change of ramps, it automatically switches to "ramp monitoring mode" when a rapid change in slope is detected. It provides centimeter-level spatial perception capabilities, excelling at capturing details of complex terrain (such as ramp curvature and abrupt slope changes), providing crucial terrain data for assessing ramp slippage risk.

[0069] In this embodiment, the combination of millimeter-wave radar, infrared camera, and lidar reduces the false alarm rate from 35% to 4.2%, and improves the target detection accuracy to 98.6% (nighttime scene). In rainy or foggy weather, the lidar point cloud is processed by bilateral filtering (spatial standard deviation σ_s=0.5, range standard deviation σ_r=0.1), reducing noise by 60% and extending the detection distance to 80 meters.

[0070] In this embodiment, a millimeter-wave radar (model: TI IWR1843, detection range 100 meters, operating frequency 77GHz) and an infrared camera (model: Sony IMX335, 0.002 Lux starlight-level night vision, resolution 1920×1080) are deployed every 50 meters inside the tunnel, connected to the edge computing unit (ECU) via a CAN bus (1Mbps baud rate). The millimeter-wave radar uses the OS-CFAR (Ordered Statistical Constant False Alarm Rate) algorithm, with the number of reference units N=32, the number of protection units G=8, and the threshold factor α=0.1. Data upload is triggered when the detected target reflection signal intensity exceeds -60 dBm. The infrared camera achieves day and night mode switching by automatically switching IR filters. When the ambient light is below 0.002 Lux, infrared supplementary lighting (wavelength 850nm, power 5W) is activated, and the frame rate is fixed at 25 FPS to ensure timing synchronization.

[0071] A LiDAR (Velodyne VLP-16, 10 Hz scanning frequency, ±15° vertical field of view, 0.1° horizontal resolution) is deployed at the top of the ramp. Point cloud data is transmitted to the ECU via Ethernet AVB (Audio-Video Bridging) protocol (100Mbps bandwidth). The LiDAR is configured with a vertical field of view of ±15° and a horizontal resolution of 0.1°. Ramp mode is triggered when a slope change rate exceeding ±0.5° / second is detected. Data from each sensor is timestamped (accuracy ±10 ms) to form a multimodal sensing data stream. The specific timestamp alignment algorithm is as follows:

[0072] 1. Embed hardware timestamps (based on GPS synchronization) during sensor data acquisition;

[0073] 2. The ECU performs local clock calibration via PTP (Precision Time Protocol);

[0074] 3. The Dynamic Time Warping (DTW) algorithm is used to align the time axis of the multi-sensor data stream, with the error controlled within ±10 ms.

[0075] Furthermore, when the ambient temperature is >60℃, the sensor-coordinated sensing unit is replaced with a high-temperature resistant component, including:

[0076] Millimeter-wave radar uses a ceramic substrate (Al2O3, thermal conductivity) The coefficient of thermal expansion is 6.5 × 10⁻⁶. -6 / ℃), the radome material was changed to polyimide (operating temperature -200℃~+260℃, dielectric constant 3.2).

[0077] The lubricating oil of the lidar motor was replaced with perfluoropolyether (PFPE), which maintains a viscosity index >150 at 85℃, reducing the motor temperature rise by 20℃.

[0078] The heatsink area of ​​the infrared camera has been increased by 50%, and a copper-aluminum composite heat pipe (thermal conductivity > 100%) has been adopted. The chip temperature is controlled at ≤80℃ by forced air cooling (wind speed 5 m / s).

[0079] In a field test at a tunnel in Chongqing (where the highest summer temperature is 65℃), the false alarm rate of this solution decreased from 8 times / day to 1.2 times / day under high-temperature conditions, and the system ran continuously for 72 hours without failure.

[0080] Furthermore, when the ambient temperature is < -20℃, the following improvements are adopted:

[0081] The millimeter-wave radar transmit power has been increased to 20 dBm, and a temperature compensation algorithm has been added (the phase offset Δφ = 0.1 rad is adjusted when ΔT = ±3℃).

[0082] The lidar point cloud data is filtered by both sides (σ_s=0.8, σ_r=0.2) to reduce ice crystal interference, and a point cloud interpolation algorithm is enabled, with a missing point filling rate of >95%.

[0083] The power of the infrared camera heating module has been increased from 5W to 15W, and a PTC ceramic heating element (response time < 10 seconds) has been adopted to ensure that the lens surface temperature is > 0℃.

[0084] In a test on a slope in Harbin (winter minimum temperature -30℃), the target detection distance was extended from 70 meters to 90 meters under low temperature conditions, the response time was stabilized within 160 ms, and the false negative rate was reduced from 25% to 5%.

[0085] Understandably, all sensors have built-in high-precision clocks and undergo unified time synchronization to ensure that data from radar, cameras, and lidar are strictly aligned in time. The edge computing unit uses intelligent algorithms to precisely match the data streams from different sensors in the time dimension. It integrates sensing data (point clouds, images) of different principles and formats into a unified "environmental situation stream" with precise spatiotemporal labels, providing an accurate and consistent data foundation for subsequent comprehensive analysis and avoiding misjudgments caused by time asynchrony.

[0086] An edge computing unit is used to receive the environmental data, perform fusion analysis, risk assessment, and make decisions based on the environmental data.

[0087] In this embodiment, the edge computing unit adopts the NVIDIA Jetson AGX Orin platform and integrates the following modules:

[0088] Multi-source data fusion module: This module uses optimized filtering algorithms to calibrate the spatial position of target point cloud information provided by millimeter-wave radar and lidar, eliminating positional errors between sensors and forming a more accurate unified target trajectory. It employs a lightweight, high-performance YOLOv8 model to perform real-time analysis of infrared images, quickly and accurately detecting "abnormal parking" events (i.e., vehicles that have been stationary for extended periods in non-parking areas).

[0089] As can be understood, this embodiment spatially registers the sparse point cloud of millimeter-wave radar with the dense point cloud of lidar using an optimized Kalman filter. Key parameters of the filter (such as process noise and observation noise covariance) are optimized offline using the Pigeon Swarm Intelligence (PIO) algorithm to achieve rapid convergence and improve accuracy. The output is a high-confidence, high-precision unified target point cloud set, with each target possessing precise three-dimensional position and velocity vectors.

[0090] As can be understood, this embodiment inputs infrared camera video frames into a lightweight modified YOLOv8 model. The model is trained using an "abnormal parking" scenario dataset and can identify vehicles and distinguish whether they are in a "driving," "compliantly parked," or "abnormally parked" state. The identified target bounding boxes are associated and matched with the point cloud targets fused in the first step, and semantic labels are assigned to the physical targets (e.g., "stationary vehicle, located in the driving lane, has been there for 35 seconds").

[0091] Please see Figure 3 The Kalman filter parameters are optimized using the Pigeon Flock Intelligent Algorithm (PI0), and the specific steps are as follows:

[0092] 1. Initialize pigeon flock parameters: Population size N=50, maximum number of iterations T=100;

[0093] 2. Design the cost function: J = ||z_k-Hx_k||², where z_k is the observed value and H is the observation matrix;

[0094] 3. Map compass operator optimization: The updated formula is X_i (t+1) = X_i (t) + ω(t) * (P_g (t) -X_i (t)) + rand (-1,1) * (X_i (t) - X_g (t)), where ω(t) is the contraction and expansion coefficient;

[0095] 4. Landmark operator optimization: Calculate the group weighted center X_c = Σ(X_i * fitness (X_i)) / Σfitness (X_i), and retain the top 50% of the best individuals.

[0096] The pigeon flocking algorithm optimizes the Kalman filter parameters in two stages: map compass operator and landmark operator. The optimized parameters are used for spatial alignment between millimeter-wave radar and lidar, improving positioning accuracy to ±0.2 meters.

[0097] Infrared images were used for target detection using a YOLOv8 model with an input size of 640×640 pixels. The model employed the CIoU loss function and was trained on a tunnel scene dataset (containing 100,000 labeled images, of which 15% were abnormal parking samples) to achieve an mAP@0.5=92.7%. The YOLOv8 backbone network incorporated the GhostConv module, replacing the first convolutional layer, reducing the number of parameters by 18% and improving inference speed by 15%. When a stationary vehicle was detected for more than 30 seconds and was not in a parking space (parking space detection was achieved through an auxiliary branch of YOLOv8 with an accuracy of 98.3%), it was classified as abnormal parking.

[0098] It is understandable that the use of multi-source data fusion modules for spatial fusion improves the accuracy of target positioning; visual recognition gives the system the ability to understand scene semantics (such as whether it is a violation), which greatly reduces the possibility of false alarms or missed alarms from a single sensor.

[0099] The risk assessment module integrates slope data and vehicle speed information from the onboard CAN bus to calculate the speed change rate Δv and trajectory offset, and adopts a dual-threshold graded intervention.

[0100] A runaway warning is triggered when a sudden change in vehicle speed Δv > 20 km / h / s and a trajectory deviation > 1.5 meters is detected. The system employs a dual-threshold decision-making mechanism.

[0101] i. Primary warning (Δv > 15 km / h / s): Activate brake precharge, and the hydraulic system pressure rises to 50 bar;

[0102] ii. Advanced warning (Δv > 20 km / h / s): Directly triggers ESP hydraulic braking, with braking deceleration ≥ 6 m / s². Trajectory deviation calculation employs a Kalman filter-based multi-target tracking algorithm, achieving a tracking accuracy of ±0.3 meters and a response time < 50 ms.

[0103] The adaptive threshold adjustment module, based on a fuzzy logic controller, dynamically adjusts the radar detection threshold and camera confidence threshold according to real-time traffic density, ambient light, and weather conditions.

[0104] In this embodiment, based on the traffic flow density model, when the tunnel traffic flow is >80 vehicles / h, the millimeter-wave radar detection threshold is increased to -55dBm, and the infrared camera false detection tolerance is reduced from 0.5 to 0.3. The input variables of the fuzzy logic controller are traffic flow density (domain of discourse: 0-200 vehicles / h), ambient light (domain of discourse: 0-1000 Lux), and weather conditions (sunny, rainy, foggy). The output variables are the millimeter-wave radar threshold adjustment value (ΔThr_radar) and the infrared camera confidence threshold (Thr_cam). The fuzzy rule base contains 27 rules, for example:

[0105] If (traffic density is high) and (light intensity is low) and (weather is foggy), then (ΔThr_radaris +10 dBm, Thr_cam is 0.3).

[0106] The fuzzy inference uses the Mamdani method, and the centroid method is used for defuzzification. The system response time is <50 ms, the false alarm rate in normal scenarios is controlled to <1 time / day, and the false alarm rate in rainy and foggy weather is allowed to be ≤3 times / day.

[0107] Please see Figure 4 Instead of using fixed parameters, the system dynamically adjusts radar detection sensitivity and image recognition confidence requirements based on real-time traffic flow, ambient lighting, and weather conditions, using fuzzy logic rules. This makes risk assessment more intelligent and realistic. In complex environments with heavy traffic and low visibility, the system automatically increases alertness (adjusting thresholds), while maintaining a more lenient approach in simpler environments to reduce interference, achieving a balance between high detection rate and low false alarm rate.

[0108] Specifically, it acquires data from light sensors and weather sensors (or infers data from video analysis) in real time. Based on the fused target list, it calculates the number of tire tracks passed per unit time in real time to obtain the current traffic density. It also acquires real-time data on vehicle speed and turn signal status of itself and nearby vehicles via the vehicle's CAN bus. LiDAR continuously analyzes the slope and rate of change to determine if the vehicle is on a slope.

[0109] The above-mentioned "traffic density", "ambient light" and "weather conditions" are used as inputs and sent to a fuzzy logic controller with 27 preset experience points.

[0110] The controller simulates human expert decision-making, outputting dynamic adjustments to the millimeter-wave radar detection threshold and camera recognition confidence threshold. For example, in situations such as heavy fog, nighttime, or heavy traffic, it automatically increases radar sensitivity (raises the threshold) and decreases tolerance for image recognition results (requiring greater certainty).

[0111] For the risk of vehicle rollover on slopes: three core indicators are considered: "real-time slope", "vehicle speed change value (Δv)", and "lateral deviation of vehicle trajectory". A dual-threshold triggering mechanism is adopted: when Δv > 15 km / h / s, it is marked as a primary risk; when Δv > 20 km / h / s and trajectory deviation > 1.5 meters, it is judged as a high-level risk.

[0112] For abnormal parking risks: the judgment is made directly based on the semantic information output ("abnormal parking" and "duration > 30 seconds"), and all risks are normalized into a comprehensive risk value of 0-1.

[0113] It is understandable that the input variables are traffic density, ambient light, and weather conditions, which are converted into fuzzy sets through fuzzification. The rule base contains 27 rules, and the Mamdani inference method is used to output millimeter-wave radar threshold adjustment values ​​and infrared camera confidence thresholds after defuzzification.

[0114] The multi-level early warning execution unit is used to initiate a progressive response from information early warning to active control based on decision instructions.

[0115] In this embodiment, the multi-level early warning execution unit includes controlling an audible and visual alarm via a LIN bus, and simultaneously pushing multi-level early warning information to a cloud platform via a 4G module.

[0116] Wherein: the multi-level early warning information includes:

[0117] Yellow alert: Triggers in-vehicle audio and visual alerts to warn the driver;

[0118] Orange alert: Based on the yellow alert, lane-level speed limit information is broadcast via DSRC, and pre-braking is performed on at-risk vehicles to achieve vehicle-road cooperative early warning;

[0119] Red Alert: Directly activate ESP to perform emergency braking, simultaneously turn on hazard warning lights, and synchronize the alert to the cloud platform via 4G network.

[0120] Specifically, the local audio-visual alarm receives commands from the ECU and emits a high-brightness flashing light and a high-decibel warning sound at the tunnel site to alert drivers of vehicles behind to the potential danger ahead. This provides a direct and strong remote visual and auditory warning to vehicles behind, giving drivers more time to react.

[0121] Furthermore, through the vehicle's internal network, dashboard warnings and buzzers are triggered to directly alert the driver in the at-risk vehicle. At higher risk levels, speed limits or warning messages are broadcast to surrounding vehicles. At the highest risk level (such as an impending collision), the system has the authority to pre-charge or automatically apply emergency braking via the vehicle's braking system, and activate hazard warning lights. This achieves a progressive safety guarantee from "alerting the driver" to "assisting the driver" and then to "system takeover in emergencies." The highest level of automatic braking is the last line of defense to prevent accidents.

[0122] Furthermore, all early warning events, equipment status, and traffic data are received via wireless network, enabling the monitoring center to perform global oversight, data analysis, and historical data review. This achieves networked and intelligent management of the devices, supporting remote monitoring, big data analysis, and continuous optimization of system strategies.

[0123] Please see Figure 5 Here is a risk assessment timing diagram for the safety early warning device provided in this embodiment, wherein:

[0124] 1. Data Input (Left Side - Sensor Array)

[0125] Millimeter-wave radar data: provides target distance and speed information at high speed with 10-millisecond intervals.

[0126] Infrared imaging: Provides visual images at a fixed frame rate of 25 FPS (frames per second) for target identification.

[0127] LiDAR point cloud: Provides high-precision 3D environmental information at a frequency of 10 Hz (i.e., 10 frames / second).

[0128] 2. Data processing core (Central-Edge Computing Unit, ECU): The ECU executes the complete process shown in the diagram every 200ms.

[0129] Data fusion and YOLOv8 inference: This is the core step in the upper path (abnormal parking detection). The ECU performs spatiotemporal alignment and fusion of data from millimeter-wave radar, infrared cameras, etc., and runs a lightweight YOLOv8 model for image recognition to determine whether there is a vehicle and whether it is in an abnormal parking state.

[0130] Slope / Vehicle Joint Calculation: This is the core step in the next path (rollback risk assessment). The ECU combines real-time slope and vehicle speed to calculate the speed change rate (Δv).

[0131] Fuzzy logic threshold adjustment: This is an intelligent module that runs continuously in the background (represented by an independent box in the figure). It dynamically adjusts the sensitivity of preceding perception modules (such as radar detection threshold and image recognition confidence) based on real-time traffic, lighting, and weather conditions to optimize overall performance.

[0132] 3. Decision Output and Early Warning Execution (Right Side)

[0133] Based on different processing paths and judgment results, the system triggers a three-level warning:

[0134] Yellow Alert: Triggered by the above path. Activated when an abnormal vehicle stoppage is detected for more than 30 seconds. The action is to trigger an audible and visual alarm to provide on-site warning (arrow pointing from the ECU to the audible and visual alarm in the diagram).

[0135] Orange alert: Triggered by the following path. Activated when calculated Δv > 15 km / h / s. This indicates the initial formation of a risk of runaway, and the system enters a high-alert state.

[0136] Red Alert: Triggered by the following path, this is a higher level of response. It activates when Δv > 20 km / h / s. This indicates a high-risk rollaway event is occurring, and the system will execute the strongest intervention (such as emergency braking).

[0137] The system performs a full-process calculation every 200ms, and achieves risk classification and early warning through a multi-stage decision tree. Data fusion and target detection take about 120ms, and risk assessment and threshold adjustment take about 80ms, ensuring that the response time for critical events is less than 200ms.

[0138] The safety early warning device provided by this invention overcomes the bottleneck of single sensors due to changes in tunnel ambient lighting and weather conditions by fusing multi-dimensional data collection, including distance, speed, visual images, and three-dimensional terrain data. It achieves reliable monitoring 24 hours a day under various weather conditions. Complex data fusion and AI analysis are performed locally with extremely low response latency, meeting the stringent real-time requirements of safety systems. Through fuzzy logic control, system parameters can be intelligently adjusted according to traffic and environmental conditions, ensuring high sensitivity while keeping the daily false alarm rate extremely low. A complete intervention chain is established from remote warning and near-end alert to direct vehicle control, realizing smooth and safe takeover under the "human-machine co-driving" concept. This effectively prevents and mitigates traffic accidents in tunnel and ramp scenarios and is suitable for safety monitoring and risk warning of intelligent connected vehicles in complex road conditions.

[0139] Please see Figure 6 The present invention also provides a safety early warning method, which is used to implement the safety early warning device provided in the above embodiments, specifically including:

[0140] Collect environmental data, including distance, speed, visual images, and three-dimensional terrain.

[0141] The specific implementation method is as follows:

[0142] Millimeter-wave radar: Responsible for detecting the distance, speed, and azimuth of moving and stationary targets at medium to long ranges. Employing advanced signal processing algorithms, it effectively filters out environmental clutter interference, reporting data only when the target's reflected signal is sufficiently strong, ensuring reliable detection. It achieves stable speed and distance measurement in adverse weather conditions such as rain, fog, and dust, providing the system with fundamental moving target data.

[0143] Infrared camera: Capable of imaging in extremely low light conditions, automatically activating infrared illumination in dimly lit tunnels or at night to acquire clear video images. Through intelligent analysis algorithms, it identifies vehicle outlines, parking status, and whether the parking space is compliant. This solves the problem of traditional cameras failing in insufficient lighting within tunnels, enabling 24 / 7 uninterrupted visual monitoring and accurate target classification (e.g., distinguishing between normal driving and abnormal parking).

[0144] LiDAR: Generates high-precision 3D point cloud maps by emitting laser beams and receiving reflections. Specifically designed to sense the slope and rate of change of ramps, it automatically switches to "ramp monitoring mode" when a rapid change in slope is detected. It provides centimeter-level spatial perception capabilities, excelling at capturing details of complex terrain (such as ramp curvature and abrupt slope changes), providing crucial terrain data for assessing ramp slippage risk.

[0145] In this embodiment, the combination of millimeter-wave radar, infrared camera, and lidar reduces the false alarm rate from 35% to 4.2%, and improves the target detection accuracy to 98.6% (nighttime scene). In rainy or foggy weather, the lidar point cloud is processed by bilateral filtering (spatial standard deviation σ_s=0.5, range standard deviation σ_r=0.1), reducing noise by 60% and extending the detection distance to 80 meters.

[0146] Receive the environmental data, perform fusion analysis and risk assessment on the environmental data, and make decisions.

[0147] In this embodiment, the target point cloud information provided by millimeter-wave radar and lidar is spatially calibrated using an optimized filtering algorithm to eliminate positional errors between sensors and form a more accurate unified target trajectory. A lightweight, high-performance YOLOv8 model is used to analyze the infrared images in real time, quickly and accurately detecting "abnormal parking" events (i.e., vehicles that have been stationary for extended periods in non-parking areas).

[0148] As can be understood, this embodiment spatially registers the sparse point cloud of millimeter-wave radar with the dense point cloud of lidar using an optimized Kalman filter. Key parameters of the filter (such as process noise and observation noise covariance) are optimized offline using the Pigeon Swarm Intelligence (PIO) algorithm to achieve rapid convergence and improve accuracy. The output is a high-confidence, high-precision unified target point cloud set, with each target possessing precise three-dimensional position and velocity vectors.

[0149] As can be understood, this embodiment inputs infrared camera video frames into a lightweight modified YOLOv8 model. The model is trained using an "abnormal parking" scenario dataset and can identify vehicles and distinguish whether they are in a "driving," "compliantly parked," or "abnormally parked" state. The identified target bounding boxes are associated and matched with the point cloud targets fused in the first step, and semantic labels are assigned to the physical targets (e.g., "stationary vehicle, located in the driving lane, has been there for 35 seconds").

[0150] Specifically, the gradient data and vehicle speed information from the onboard CAN bus are integrated to calculate the speed change rate Δv and the trajectory offset, and a dual-threshold hierarchical intervention is adopted.

[0151] A runaway warning is triggered when a sudden change in vehicle speed Δv > 20 km / h / s and a trajectory deviation > 1.5 meters is detected. The system employs a dual-threshold decision-making mechanism.

[0152] i. Primary warning (Δv > 15 km / h / s): Activate brake precharge, and the hydraulic system pressure rises to 50 bar;

[0153] ii. Advanced warning (Δv > 20 km / h / s): Directly triggers ESP hydraulic braking, with braking deceleration ≥ 6 m / s². Trajectory deviation calculation employs a Kalman filter-based multi-target tracking algorithm, achieving a tracking accuracy of ±0.3 meters and a response time < 50 ms.

[0154] Furthermore, based on a fuzzy logic controller, the radar detection threshold and camera confidence threshold are dynamically adjusted according to real-time traffic density, ambient light, and weather conditions.

[0155] In this embodiment, based on the traffic flow density model, when the tunnel traffic flow is >80 vehicles / h, the millimeter-wave radar detection threshold is increased to -55dBm, and the infrared camera false detection tolerance is reduced from 0.5 to 0.3. The input variables of the fuzzy logic controller are traffic flow density (0-200 vehicles / h), ambient light (0-1000 Lux), and weather conditions (sunny, rainy, foggy), and the output variables are the millimeter-wave radar threshold adjustment value (ΔThr_radar) and the infrared camera confidence threshold (Thr_cam).

[0156] Based on the decision-making instructions, a progressive response is initiated, from information early warning to proactive control.

[0157] The audible and visual alarm is controlled via LIN bus, and multi-level early warning information is pushed to the cloud platform via 4G module.

[0158] Wherein: the multi-level early warning information includes:

[0159] Yellow alert: Triggers in-vehicle audio and visual alerts to warn the driver;

[0160] Orange alert: Based on the yellow alert, lane-level speed limit information is broadcast via DSRC, and pre-braking is performed on at-risk vehicles to achieve vehicle-road cooperative early warning;

[0161] Red Alert: Directly activate ESP to perform emergency braking, simultaneously turn on hazard warning lights, and synchronize the alert to the cloud platform via 4G network.

[0162] Specifically, the local audio-visual alarm receives commands from the ECU and emits a high-brightness flashing light and a high-decibel warning sound at the tunnel site to alert drivers of vehicles behind to the potential danger ahead. This provides a direct and strong remote visual and auditory warning to vehicles behind, giving drivers more time to react.

[0163] Furthermore, through the vehicle's internal network, dashboard warnings and buzzers are triggered to directly alert the driver in the at-risk vehicle. At higher risk levels, speed limits or warning messages are broadcast to surrounding vehicles. At the highest risk level (such as an impending collision), the system has the authority to pre-charge or automatically apply emergency braking via the vehicle's braking system, and activate hazard warning lights. This achieves a progressive safety guarantee from "alerting the driver" to "assisting the driver" and then to "system takeover in emergencies." The highest level of automatic braking is the last line of defense to prevent accidents.

[0164] Furthermore, all early warning events, equipment status, and traffic data are received via wireless network, enabling the monitoring center to perform global oversight, data analysis, and historical data review. This achieves networked and intelligent management of the devices, supporting remote monitoring, big data analysis, and continuous optimization of system strategies.

[0165] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 7As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the security warning methods described in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0166] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0167] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0168] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0169] This invention also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps of any of the security warning methods described in the above embodiments. The computer-readable storage medium can be volatile or non-volatile.

[0170] This invention also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-mentioned security warning method.

[0171] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0172] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0173] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0174] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0175] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0176] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0177] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0178] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0179] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0180] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A safety early warning device, characterized in that, It includes: A sensor-coordinated sensing unit is used to collect environmental data, including distance, speed, visual images, and three-dimensional terrain. An edge computing unit is used to receive the environmental data, perform fusion analysis, risk assessment, and make decisions based on the environmental data; The multi-level early warning execution unit is used to initiate a progressive response from information early warning to active control based on decision instructions.

2. The safety early warning device according to claim 1, characterized in that, The sensor collaborative sensing unit includes: millimeter-wave radar, infrared camera, and lidar; Furthermore, the aforementioned lidar is used to generate a high-precision three-dimensional point cloud map by emitting a laser beam and receiving reflections, and to automatically switch to slope monitoring mode when a rapid change in slope is detected by sensing the slope and its rate of change.

3. The safety early warning device according to claim 2, characterized in that, When the ambient temperature is >60℃, the lubricating oil of the lidar motor is perfluoropolyether; when the ambient temperature is <-20℃, the lidar point cloud data is processed by bilateral filtering and a point cloud interpolation algorithm is used.

4. The safety early warning device according to claim 2, characterized in that, The edge computing unit includes: Multi-source data fusion module: fuses LiDAR data and millimeter-wave radar data, and uses infrared images to perform target detection through the YOLOv8 model; Risk assessment module: It integrates slope data and vehicle speed information from the vehicle CAN bus to calculate the speed change rate Δv and trajectory offset, and adopts a dual-threshold graded intervention. Furthermore, when Δv > 15 km / h / s, a primary warning is issued, and the braking system is pre-charged. When Δv > 20 km / h / s and the deviation > 1.5 meters, a high-level warning is issued, directly triggering ESP emergency braking; Adaptive threshold adjustment module: Based on a fuzzy logic controller, it dynamically adjusts the radar detection threshold and camera confidence threshold according to real-time traffic density, ambient light, and weather conditions.

5. The safety early warning device according to claim 4, characterized in that, The infrared image is used for target detection using the YOLOv8 model, including: The backbone network of the YOLOv8 model introduces the GhostConv module, replacing the first convolutional layer; Furthermore, in tunnel scenarios, when a stationary vehicle is detected to have exceeded a set time and is not located in a parking space, it is determined to be an abnormal parking situation.

6. The safety early warning device according to claim 1, characterized in that, The multi-level early warning execution unit includes: Yellow alert: Triggers in-vehicle audio and visual alerts to warn the driver; Orange alert: Triggers in-vehicle audio and visual alerts to warn the driver, broadcasts lane-level speed limit information, and pre-charges the brakes on at-risk vehicles; Red alert: Directly activates ESP to perform emergency braking, simultaneously turns on hazard warning lights, and synchronizes the alert to the cloud platform.

7. The safety early warning device according to claim 1, characterized in that, The sensor collaborative sensing unit includes: Hardware timestamps are embedded during sensor data acquisition; The ECU performs local clock calibration via PTP; A dynamic time warping algorithm is used to align the time axis of multi-sensor data streams, with the error controlled within 10ms.

8. A safety early warning method, characterized in that, For implementing the apparatus according to any one of claims 1-7, comprising: Collect environmental data, including: distance, speed, visual images, and three-dimensional terrain; Receive the environmental data, perform fusion analysis and risk assessment on the environmental data, and make decisions; Based on the decision-making instructions, a progressive response is initiated, from information early warning to proactive control.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in claim 8.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method as described in claim 8.

Images