Data processing method, processing system and automobile data recorder

Abstract

The invention discloses a data processing method, a processing system and a formed recorder. The data processing method is realized based on an automobile data recorder, the automobile data recorder is installed on the electric bicycle, and the data processing method comprises the following accident early warning steps: when the running of the electric bicycle is identified, obtaining video data of the automobile data recorder. Detecting a first target object and a second target object in the video data, and respectively generating a corresponding first target detection frame and a corresponding second target detection frame; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle. And adopting a multi-target tracking algorithm to track the first target object and the second target object respectively. And respectively predicting tracks of the first target object and the second target object by adopting a track prediction model to obtain a first prediction track and a second prediction track. And calculating whether an accident conflict point exists between the first prediction trajectory and the second prediction trajectory.

Description

Technical Field

[0001] The present invention relates to the field of driving recorder data processing, and in particular to a data processing method, a processing system and a driving recorder. Background Art

[0002] As the number of electric bicycles continues to grow, traffic accidents involving electric bicycle riders are showing a rapid growth trend.

[0003] At present, when an electric bicycle is involved in a serious traffic accident, the electric bicycle and the owner do not have a good and timely active warning method, especially on remote roads, which may result in the accident not being handled and rescued in time. If the owner is unconscious, it is even more insufficient to provide effective warning and safety protection. Summary of the invention

[0004] In order to overcome the above technical defects, the present invention provides a data processing method, a processing system and a driving recorder.

[0005] In order to solve the above problems, the present invention is implemented according to the following technical solutions:

[0006] In a first aspect, the present invention provides a data processing method, which is implemented based on a driving recorder, wherein the driving recorder is installed on an electric bicycle, and the data processing method includes the following accident warning steps:

[0007] When the electric bicycle is identified to be running, the video data of the driving recorder is obtained;

[0008] Detecting a first target object and a second target object in the video data, and generating corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle;

[0009] Using a multi-target tracking algorithm to track the first target object and the second target object respectively;

[0010] Using a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory;

[0011] Calculating whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory;

[0012] When there is an accident conflict point, generate road accident warning information;

[0013] Send accident warning information and video data to the target device.

[0014] Preferably, the data processing method includes the following vehicle abnormality warning steps:

[0015] Acquire the driving trajectory data of the electric bicycle in real time, wherein the driving trajectory data includes position information, speed information and direction information;

[0016] Based on the driving trajectory data, identifying whether the electric bicycle has an abnormal speed event, an abnormal direction event, and an abnormal trajectory event;

[0017] When at least one of an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is identified, the abnormal vehicle video data of the vehicle before and after 30 seconds of the driving recorder is intercepted;

[0018] Abnormal speed events, abnormal direction events, abnormal trajectory events and abnormal vehicle video data are sent to the target device.

[0019] Preferably, the data processing method comprises the following vehicle dumping identification steps:

[0020] When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained;

[0021] Detecting a third target object in the image data and generating a corresponding third target detection frame;

[0022] Compare the image data with the third target detection frame with the template image, and identify whether the electric bicycle has fallen over according to the position relationship of the third target detection frame in the template image;

[0023] When an electric bicycle is identified to have fallen over, a vehicle falling over alarm message is generated;

[0024] Send vehicle tipping alarm information and image data to the target device.

[0025] Preferably, the data processing method comprises the following vehicle dumping identification steps:

[0026] When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained;

[0027] detecting a third target object in the image data;

[0028] identifying an outer contour of a third target object;

[0029] Calculating a target area of ​​the third target object based on an outer contour of the third target object;

[0030] Calculating a percentage value of the target area to the area of ​​the image data;

[0031] When the percentage value is greater than a preset threshold, a vehicle tipping alarm message is generated;

[0032] Send vehicle tipping alarm information and image data to the target device.

[0033] In a second aspect, the present invention provides a data processing system, the data processing system is implemented based on a driving recorder, the driving recorder is installed on an electric bicycle, and the data processing system includes:

[0034] An acquisition module, which is used to acquire video data of a driving recorder when recognizing that an electric bicycle is running;

[0035] A detection module, which is used to detect a first target object and a second target object in the video data, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle;

[0036] A tracking module, which is used to track the first target object and the second target object respectively using a multi-target tracking algorithm;

[0037] A trajectory prediction module, which is used to use a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory;

[0038] A calculation module, which is used to calculate whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory;

[0039] A generation module, which is used to generate road accident warning information when there is an accident conflict point;

[0040] The sending module sends the accident warning information and video data to the target device.

[0041] Preferably, the data processing system is also used to perform the following vehicle abnormality warning steps:

[0042] Acquire the driving trajectory data of the electric bicycle in real time, wherein the driving trajectory data includes position information, speed information and direction information;

[0043] Based on the driving trajectory data, identifying whether the electric bicycle has an abnormal speed event, an abnormal direction event, and an abnormal trajectory event;

[0044] When at least one of an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is identified, the abnormal vehicle video data of the vehicle before and after 30 seconds of the driving recorder is intercepted;

[0045] Abnormal speed events, abnormal direction events, abnormal trajectory events and abnormal vehicle video data are sent to the target device.

[0046] Preferably, the data processing system is also used to perform the following vehicle dumping identification steps:

[0047] When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained;

[0048] Detecting a third target object in the image data and generating a corresponding third target detection frame;

[0049] Compare the image data with the third target detection frame with the template image, and identify whether the electric bicycle has fallen over according to the position relationship of the third target detection frame in the template image;

[0050] When an electric bicycle is identified to have fallen over, a vehicle falling over alarm message is generated;

[0051] Send vehicle tipping alarm information and image data to the target device.

[0052] Preferably, the data processing system is also used to perform the following vehicle dumping identification steps:

[0053] When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained;

[0054] detecting a third target object in the image data;

[0055] identifying an outer contour of a third target object;

[0056] Calculating a target area of ​​the third target object based on an outer contour of the third target object;

[0057] Calculating a percentage value of the target area to the area of ​​the image data;

[0058] When the percentage value is greater than a preset threshold, a vehicle tipping alarm message is generated;

[0059] Send vehicle tipping alarm information and image data to the target device.

[0060] In a third aspect, the present invention further provides a driving recorder, wherein the driving recorder is provided with the data processing system described in the second aspect.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention provides a data processing method, which is implemented based on a driving recorder, and the driving recorder is installed on an electric bicycle. The data processing method includes the following accident warning steps: when the electric bicycle is identified to be running, the video data of the driving recorder is obtained. The first target object and the second target object in the video data are detected, and the corresponding first target detection frame and the second target detection frame are generated respectively; wherein, the first target object is a non-motor vehicle, and the second target object is a motor vehicle. A multi-target tracking algorithm is used to track the first target object and the second target object respectively. The trajectories of the first target object and the second target object are predicted respectively by a trajectory prediction model, and a first predicted trajectory and a second predicted trajectory are obtained. Calculate whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory. When there is an accident conflict point, generate road accident warning information. Send the accident warning information and video data to the target device.

[0063] The present invention uses the video data obtained by the driving recorder, combined with the multi-target tracking algorithm and trajectory prediction model, to predict the potential conflict points between non-motor vehicles and motor vehicles, thereby providing early warning of possible traffic accidents. It is crucial to provide timely warning and rescue after the accident occurs. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings, wherein:

[0065] Figure 1 is a three-dimensional schematic diagram of a mounting bracket of the present invention;

[0066] Figure 2 It is a schematic diagram of the assembly of a mounting bracket of the present invention;

[0067] Figure 3 is a cross-sectional schematic diagram of the second clamping arm of the present invention;

[0068] Figure 4 It is a flow chart of a data processing method of the present invention;

[0069] In the figure:

[0070] 10-Connector;

[0071] 20-first clamping arm, 21-clamping claw;

[0072] 30-second clamping arm, 31-counterweight;

[0073] 40- rubber parts;

[0074] 50 - placement board, 51 - flexible pad, 52 - plastic base. DETAILED DESCRIPTION

[0075] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0076] At present, when an electric bicycle is involved in a serious traffic accident, the electric bicycle and the owner do not have a good and timely active warning method, especially on remote roads, which may result in the accident not being handled and rescued in time. If the owner is unconscious, it is even more insufficient to provide effective warning and safety protection.

[0077] refer to Figure 4 , an embodiment of the present invention provides a flow chart of a data processing method. The present invention provides a data processing technology, which can predict the potential conflict points between non-motor vehicles and motor vehicles through the video data obtained by the driving recorder, combined with a multi-target tracking algorithm and a trajectory prediction model, so as to warn of possible traffic accidents in advance. It is very important to help timely warning and rescue after the accident.

[0078] The data processing method of the present invention can be executed by a data processing system, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device, such as a driving recorder. Figure 1 As shown, the method includes:

[0079] S100: When it is recognized that the electric bicycle is running, video data of the driving recorder is obtained.

[0080] S200: Detect a first target object and a second target object in video data, and generate corresponding first target detection frames and second target detection frames, respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle.

[0081] S300: using a multi-target tracking algorithm to track the first target object and the second target object respectively.

[0082] S400: using a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory.

[0083] S500: Calculate whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory.

[0084] S600: When there is an accident conflict point, generate road accident warning information.

[0085] S700: Send the accident warning information and video data to the target device.

[0086] When a serious traffic accident occurs, especially at night or on remote roads, timely warning and rescue are crucial. Your data processing method can generate warning information and send it to the target device when the accident conflict point is detected, which helps to notify rescue personnel in time and speed up the rescue response. Specifically, the present invention describes each step of the data processing method in detail.

[0087] S100: When it is recognized that the electric bicycle is running, video data of the driving recorder is obtained.

[0088] In the present invention, a driving recorder is installed on the electric bicycle. The recorder has a video acquisition function and can record image data during driving in real time. The driving recorder captures road conditions through a built-in camera.

[0089] Preferably, a driving recorder with high-definition video recording function is installed on the electric bicycle, and the recorder should have the ability to shoot at night to adapt to different lighting conditions.

[0090] In a specific implementation, the software built into the driving recorder can automatically identify the start-up and driving status of the electric bicycle, for example, by detecting the speed change of the electric bicycle or by connecting with the electronic control unit (ECU) of the electric bicycle to obtain the driving signal.

[0091] Specifically, real-time video data is obtained from the dashcam installed on the electric bicycle. This is the basis of the entire data processing process, because without video data, subsequent target detection and tracking cannot be performed. The acquisition of video data can provide raw materials for subsequent target detection and tracking, which is the basis of the entire data processing process.

[0092] S200: Detect a first target object and a second target object in video data, and generate corresponding first target detection frames and second target detection frames, respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle.

[0093] In this step, different types of traffic participants can be accurately identified and distinguished, providing precise target information for subsequent tracking and trajectory prediction.

[0094] In a specific implementation, deep learning algorithms such as YOLO (You Only Look Once) or SSD (Single Shot MultiBox Detector) can be used to analyze the video stream in real time to identify non-motor vehicles and motor vehicles. After detecting the target object, the algorithm will mark the location of the target object in the video frame and generate a bounding box to identify these objects. The system will classify the detected target objects into non-motor vehicles (first target object) and motor vehicles (second target object) according to the preset classification criteria.

[0095] In a specific implementation, the target detection model is a Yolov5s model, which has identified the first target object, the second target object, and the third target object.

[0096] S300: using a multi-target tracking algorithm to track the first target object and the second target object respectively.

[0097] In this step, the multi-target tracking algorithm can provide information about the motion of target objects over time, which is crucial for understanding traffic flow dynamics and predicting potential conflicts.

[0098] In the specific implementation, select a suitable multi-target tracking algorithm, such as DeepSORT or MOT (Multi-Object Tracking) algorithm, which can handle problems such as target occlusion, intersection and reappearance. The algorithm will track each target object and maintain the continuity of tracking even when the target temporarily leaves the field of view or reappears after being occluded. The system will update the position and motion status of each target object in real time to provide data support for trajectory prediction.

[0099] In a specific implementation, during the target detection process, pedestrians, non-motor vehicles and motor vehicles in each frame will be detected. In order to unify the information of the same target, it is necessary to give the same target a constant ID in different frames. The DeepSORT multi-target tracking algorithm is used, which adds

[0100] DeepAssociationMetric and appearance image information are used to solve the problem of tracking occluded targets for a long time. It is a tracking framework based on deep appearance feature model and motion information model.

[0101] S400: using a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory.

[0102] In a specific implementation, machine learning or deep learning methods, such as recurrent neural networks (RNNs) or long short-term memory networks (LSTMs), are used to predict the future position and speed of each target object based on historical trajectory data.

[0103] In the specific implementation, the LSTM (Long Short-Term Memory) network prediction model is used to predict the future position and speed of each target object based on the historical trajectory data. The position and speed data obtained by the tracking algorithm are input into the trajectory prediction model. The model will output the future trajectory of each target object, namely the first predicted trajectory (non-motor vehicle) and the second predicted trajectory (motor vehicle).

[0104] In a specific implementation, the trajectory prediction model uses a Kalman filter, and may also use an extended Kalman filter predictor.

[0105] In a specific implementation, when the target object is detected, the trajectory of the object is predicted by a Kalman filter, and then the trajectory data of the predicted frame is combined and matched with the trajectory detections of the current frame using the Hungarian algorithm, and then the Kalman filter is updated.

[0106] Specifically, the DeepSORT algorithm passes the interaction ratio of the target detection box obtained by the YOLO algorithm and the previously predicted target tracking box as a combination into the Hungarian algorithm, and uses linear allocation to associate the IDs of each frame. It can also add the target's appearance and position information in the process of inter-frame matching calculation to achieve correct matching when the target object is occluded but reappears later, reducing the frequent jumps of the target ID and achieving the purpose of continuous tracking.

[0107] S500: Calculate whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory.

[0108] The present invention calculates whether the trajectories of non-motor vehicles and motor vehicles intersect or approach each other, i.e., accident conflict points, by analyzing the predicted trajectories. This step can identify potential collision risks and provide a basis for early warning.

[0109] In the specific implementation, taking the road intersection without traffic signals as an example, collision problems may occur during the dynamic interaction between electric bicycles and non-motor vehicles, electric bicycles and motor vehicles, etc. By predicting whether there are accident conflict points between the trajectories. Specifically, the predicted trajectory of the target object can be matched to the background of the image data (calibrating points along the ground in the image and forming a predicted trajectory line), and then judging whether there are overlapping intersection points between the predicted trajectories to judge whether there are accident conflict points.

[0110] S600: When there is an accident conflict point, generate road accident warning information.

[0111] In a specific implementation, once an accident conflict point is detected, the system will trigger the warning logic and generate warning information. The warning information includes the location, time, type of target object involved, etc. of the conflict point. Preferably, the system will generate an easy-to-understand warning information, which can be in the form of visual (such as screen display), auditory (such as sound alarm) or tactile (such as vibration).

[0112] In a specific implementation, the road accident warning information includes the accident conflict level, the road location information, and the location information of the driving recorder. Specifically, the accident conflict level can be calibrated according to the conflict relationship between the target objects, such as the first-level accident conflict level between electric bicycles and non-motor vehicles; the second-level accident conflict level between electric bicycles and motor vehicles.

[0113] S700: Send the accident warning information and video data to the target device.

[0114] In specific implementation, through wireless communication technology, such as 4G / 5G network, the warning information and related video data are transmitted in real time to the target device, such as the smart phone of the owner's family, or the monitoring system of the traffic management center. This step ensures that the warning information can be conveyed to the personnel who need to take action in time, improving the accident response and timely rescue.

[0115] In a preferred implementation, the data processing method further includes the following vehicle early warning step:

[0116] S10: Acquire the driving trajectory data of the electric bicycle in real time, wherein the driving trajectory data includes position information, speed information and direction information.

[0117] In a preferred implementation, the driving recorder is connected to the control system of the electric bicycle to obtain relevant data information of the electric bicycle in real time. Sensors such as GPS, speed sensor and gyroscope are installed on the electric bicycle to obtain position, speed and direction information in real time. Through these sensors, the system can collect the driving trajectory data of the electric bicycle in real time, including longitude, latitude, speed (such as kilometers per hour), acceleration, direction (such as angle or compass direction), etc.

[0118] S20: Based on the driving trajectory data, identifying whether an abnormal speed event, an abnormal direction event, and an abnormal trajectory event occur in the electric bicycle.

[0119] In a specific implementation, threshold setting is required: according to the normal driving parameters of the electric bicycle (such as speed, direction change rate, etc.), the thresholds of abnormal speed, abnormal direction and abnormal trajectory are set. The data processing system will monitor the driving trajectory data in real time according to the preset thresholds, and once an abnormal change in speed, direction or trajectory is detected, it will be identified as an abnormal event.

[0120] In a specific implementation, speed anomaly detection: a normal speed threshold is set, and when the speed exceeds the threshold, it is identified as a speed anomaly event.

[0121] Direction anomaly detection: Set a normal direction change angle threshold. When the instantaneous direction change exceeds the change angle threshold, it is identified as a direction anomaly event.

[0122] Trajectory Anomaly Detection: Identifies sudden changes in path or deviations from the normal driving trajectory by analyzing the continuity and smoothness of the driving trajectory.

[0123] By identifying abnormal events, safety risks can be discovered in a timely manner, and accidents can be quickly confirmed. When an accident does occur, timely rescue can be carried out.

[0124] S30: When at least one of an abnormal speed event, an abnormal direction event and an abnormal trajectory event is identified, abnormal vehicle video data of 30 seconds before and after the driving recorder is intercepted.

[0125] In specific implementation, once any abnormal event is detected, the system will automatically capture the dashcam video data 30 seconds before and after the abnormal event occurs, providing intuitive video evidence for the abnormal event, facilitating confirmation of the accident, post-event analysis, and liability determination, etc.

[0126] S40: Sending the abnormal speed event, abnormal direction event, abnormal trajectory event and abnormal vehicle video data to the target device.

[0127] In a specific implementation, the target device can receive abnormal event information and video data in real time and respond in a timely manner.

[0128] The vehicle early warning step of the present invention aims to timely detect the risk of an accident that has occurred, confirm the accident and provide rescue in a timely manner by real-time monitoring and identifying abnormal events in speed, direction and trajectory.

[0129] Furthermore, the data processing method of the present invention includes a vehicle tipping identification step. Under normal circumstances, the vehicle maintains a stable and upright driving state during normal driving. The vehicle will only tip over when the owner has an accident or suffers a traffic accident. To this end, timely warnings are issued by identifying whether the vehicle has tipped over. The method can timely identify the tipping state of the electric bicycle and quickly generate alarm information, providing valuable time for rescue. Through real-time analysis and comparison of image data, the accuracy of tipping identification is improved and the possibility of false alarms is reduced. The present invention combines multiple sensor data and image processing technology to provide a comprehensive solution to identify and respond to abnormal conditions of electric bicycles.

[0130] Specifically, the present invention provides the following two implementation modes:

[0131] The first specific implementation method: the data processing method includes the following vehicle dumping identification steps:

[0132] S1000: When an accident warning message is generated, or at least one of an abnormal speed event, an abnormal direction event, and an abnormal trajectory event occurs, image data captured by the driving recorder 30 seconds later is acquired.

[0133] Specifically, after generating an accident warning message or identifying abnormal speed, direction, or trajectory events, the system will instruct the driving recorder to capture an image data every 30 seconds.

[0134] Specifically, the time interval after 30 seconds is to reduce misjudgment. If there is no other abnormality with the electric bicycle, or it is accidentally overturned, the owner will generally straighten it within 30 seconds to minimize misjudgment.

[0135] S2000: Detect a third target object in the image data and generate a corresponding third target detection frame.

[0136] In the specific implementation, the above-mentioned related technical description can be referred to. In the present invention, the third target object can be one of a building, the sky, and the ground. A bounding box is generated around the identified target object to clarify the position of the target object in the image.

[0137] S3000: Compare the image data with the third target detection frame with the template image, and identify whether the electric bicycle has fallen over based on the positional relationship of the third target detection frame in the template image.

[0138] By presetting a template image, which represents the normal upright state of the electric bicycle, the real-time image data is compared with the template image, and the position relationship of the third target detection frame in the template image is analyzed to determine whether the posture of the electric bicycle is normal.

[0139] S4000: When it is identified that the electric bicycle has tipped over, a vehicle tipping alarm message is generated.

[0140] In the specific implementation, the tipping judgment logic: according to the comparison result between the third target detection frame and the template image, combined with the preset tipping judgment logic, it is identified whether the electric bicycle has tipped over. For example, in the template image, the sky should be located directly above the template image. When the electric bicycle tips over, the sky may be on the left or right side of the image, so as to judge whether the electric bicycle has tipped over.

[0141] S5000: Send the vehicle tipping alarm information and image data to the target device.

[0142] In specific implementation, once the system identifies that an electric bicycle has fallen, it will automatically generate a vehicle fall alarm message, including the time, location and possible cause of the fall. Alarm message content: The alarm message should contain sufficient details so that the owner's relatives, rescuers or the owner can respond quickly and take appropriate actions.

[0143] In a second specific implementation, the data processing method includes the following vehicle dumping identification steps:

[0144] S1000: When an accident warning message is generated, or at least one of an abnormal speed event, an abnormal direction event, and an abnormal trajectory event occurs, image data captured by the driving recorder 30 seconds later is acquired.

[0145] Specifically, by acquiring 30 seconds of image data, the situation within a period of time after the abnormal event occurs can be captured, which is helpful for comprehensive analysis. After generating accident warning information or identifying abnormal events in speed, direction, and trajectory, the system will instruct the driving recorder to capture an image data every 30 seconds.

[0146] Specifically, the time interval after 30 seconds is to reduce misjudgment. If there is no other abnormality with the electric bicycle, or it is accidentally overturned, the owner will generally straighten it within 30 seconds to minimize misjudgment.

[0147] S2000: Detect a third target object in the image data.

[0148] In the specific implementation, the above-mentioned related technical description can be referred to. Accurately identifying and locating the target object in the image provides key information for subsequent dumping identification. In this embodiment, the third target object is preferably the ground.

[0149] S3000: Identify the outer contour of a third target object.

[0150] In a specific implementation, an image processing technique (such as the Canny edge detection algorithm) is used to identify the outer contour of the target object. The extracted contour is analyzed to determine the shape and size of the target object.

[0151] S4000: Calculate a target area of ​​the third target object based on the outer contour of the third target object.

[0152] In a specific implementation, a function in an image processing library (such as the contourArea function in OpenCV) is used to calculate the area of ​​the target object, and the calculated target area is recorded for subsequent percentage calculation.

[0153] By calculating the area, the size of the target object can be quantified, providing a numerical basis for dumping identification. Area calculation provides an accurate measure of the size of the target object, which helps to improve the accuracy of dumping identification.

[0154] S5000: Calculate the percentage value of the target area to the area of ​​the image data.

[0155] In a specific implementation, by calculating the percentage value, the size of the target object relative to the entire image can be analyzed, which helps to identify significant dumping events. The percentage value provides a standardized method to compare the size of target objects in different images, which improves the versatility of the method. When an electric bicycle dumps, the ground often occupies a large part of the area. For this reason, the image ratio of the target object in the dashcam is identified to identify whether the vehicle has dumped.

[0156] S6000: When the percentage value is greater than a preset threshold, a vehicle tipping alarm message is generated.

[0157] In the specific implementation, it is possible to identify the dumping state of an electric bicycle in real time and respond quickly. Rapid dumping recognition reduces the delay from the occurrence of an event to the issuance of an alarm, and improves the efficiency of emergency response.

[0158] S7000: Send the vehicle tipping alarm information and image data to the target device.

[0159] In specific implementation, once the system identifies that an electric bicycle has fallen, it will automatically generate a vehicle fall alarm message, including the time, location and possible cause of the fall. Alarm message content: The alarm message should contain sufficient details so that the owner's relatives, rescuers or the owner can respond quickly and take appropriate actions.

[0160] The present invention also provides a data processing system, which is implemented based on a driving recorder, and the driving recorder is installed on an electric bicycle. The data processing system includes:

[0161] An acquisition module, which is used to acquire video data of a driving recorder when recognizing that an electric bicycle is running;

[0162] A detection module, which is used to detect a first target object and a second target object in the video data, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle;

[0163] A tracking module, which is used to track the first target object and the second target object respectively using a multi-target tracking algorithm;

[0164] A trajectory prediction module, which is used to use a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory;

[0165] A calculation module, which is used to calculate whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory;

[0166] A generation module, which is used to generate road accident warning information when there is an accident conflict point;

[0167] The sending module sends the accident warning information and video data to the target device.

[0168] Specifically, the data processing system of the present invention is also used to implement and execute the vehicle abnormality warning step and the vehicle tipping identification step in the above data processing method.

[0169] The present invention also provides a driving recorder, which is used to deploy a data processing system to implement the above-mentioned data processing method. Specifically, the data processing method is executed and implemented by a processor of the driving recorder of the electric bicycle.

[0170] The driving recorder of the present invention comprises a housing and a mounting bracket, wherein the housing is equipped with circuit board modules such as a camera and a processor.

[0171] like Figure 1As shown, the product of the present invention has better versatility and practicality, and can be used on motorcycles, electric bicycles and cars respectively. Specifically, the mounting bracket includes a connecting member 10, a first clamp arm 20, a second clamp arm 30 and an elastic member. The connecting member is used to connect the camera; the first clamp arm is connected to the connecting member; the second clamp arm is hinged to the first clamp arm, and the second clamp arm is located below the first clamp arm. The second clamp arm has a planar structure; the elastic member is connected to the first clamp arm and the second clamp arm, and the elastic member is used to drive the first clamp arm and the second clamp arm to clamp each other. Among them, the mounting bracket can form a clamping mechanism through the first clamp arm and the second clamp arm for clamping on an external carrier; the mounting bracket can also be stably placed on the external carrier through the planar structure of the second clamp arm.

[0172] In the present technology, the connecting piece is the basic part of the mounting bracket, which is used to connect the camera. The elastic piece is used to provide elastic force so that the first clamping arm and the second clamping arm can be clamped together, thereby being fixed to the relevant part of the motorcycle or electric bicycle.

[0173] In a specific implementation, the connecting member is hinged to the first clamp arm, and the connecting member can swing around the hinge. The connecting member is a T-shaped structure, and the first clamp arm is provided with a mounting groove structure adapted to the connecting member.

[0174] This patent forms a clamping mechanism by cooperating with the first clamping arm and the second clamping arm, and under the action of the elastic member, the clamping mechanism can stably clamp on the handlebars of motorcycles and electric bicycles, so as to meet the use of motorcycles / electric bicycles. In order to adapt to the use of family cars, the planar structure of the second clamping arm allows it to be stably placed on the car interior, such as the co-pilot dashboard.

[0175] In a specific implementation, the first clamp arm, the second clamp arm and the connecting piece are all plastic parts molded by plastic injection molding.

[0176] The main bodies of the first clamp arm and the second clamp arm are relatively flat. The first clamp wall has two bent clamp claws 21, which are arranged relatively spaced apart, and the ends of the clamp claws extend toward the second clamp arm. The bent clamp claws can form a claw-shaped clamping mechanism, which can adapt to more external carriers and improve the practicality of the mounting bracket.

[0177] In a preferred embodiment, a first rubber piece is disposed at the end of the clamping jaw; a second rubber piece is disposed at the end of the second clamping arm adjacent to the clamping jaw; wherein both the first rubber piece and the second rubber piece are rubber blocks with elliptical cross-sections.

[0178] Through this design, the oval cross-section rubber part can provide better friction when compressed, which helps the clamping jaws to grip the structure of the motorcycle or electric bicycle more firmly and prevent it from slipping during use. On the other hand, the rubber part can adapt to different contact surface shapes and maintain good contact.

[0179] In a specific implementation of the present invention, the elastic member is a torsion spring. It is understandable that the first clamp arm and the second clamp arm of the present invention are like clamps, and the torsion spring provides a stable clamping force, so that the first clamp arm and the second clamp arm can achieve a clamping effect. Specifically, using a torsion spring, a first clamp arm and a second clamp arm to form a clamping mechanism is common knowledge in the art and can be easily implemented by those skilled in the art, so no further explanation is given here.

[0180] In a preferred embodiment, the second clamping arm includes a plastic shell and a plurality of counterweights disposed in the plastic shell.

[0181] Among them, the counterweight block increases the weight of the mounting bracket itself, and is mainly used to stabilize the center of gravity of the mounting bracket and increase the weight of the mounting bracket itself to ensure stability when placed on the car interior so that it can remain stable.

[0182] In a preferred implementation, the mounting bracket also includes a placement plate 50, which has a plurality of magnets built therein and is used to be placed on an external carrier; wherein the counterweight block of the second clamping arm is an iron block, and the second clamping arm can be magnetically connected to the placement plate through the counterweight block.

[0183] The magnetic connection allows the second clamp arm and the placement plate to be quickly and easily installed or removed without complicated alignment or fastening operations, which improves the convenience of use. The placement plate is designed to be used in a car, where the placement plate serves as a base for the mounting bracket to provide stable placement. With the placement plate as a base, the placement plate has a larger area for more stable placement on the car interior. The second clamp arm is then magnetically connected to the placement plate to stably place the entire mounting bracket.

[0184] In a specific implementation, the placement plate includes a flexible pad 51 and a plastic base 52 fixedly connected to the flexible pad; wherein the magnet is disposed in the plastic base.

[0185] In the present technology, the flexible pad can adapt to different carrier surfaces, including curved or uneven surfaces, to improve applicability. The surface of the flexible pad usually has good friction properties, which can increase the friction between the placement plate and the carrier to prevent the camera from sliding or shifting.

[0186] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A data processing method, characterized in that: Based on a driving recorder installed on an electric bicycle, the data processing method includes the following accident warning steps: When the electric bicycle is identified to be running, the video data of the driving recorder is obtained; Detecting a first target object and a second target object in the video data, and generating corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle; Using a multi-target tracking algorithm to track the first target object and the second target object respectively; Using a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory; Calculating whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory; When there is an accident conflict point, generate road accident warning information; Send accident warning information and video data to the target device.

2. A data processing method according to claim 1, characterized in that: The data processing method includes the following vehicle abnormality warning steps: Acquire the driving trajectory data of the electric bicycle in real time, wherein the driving trajectory data includes position information, speed information and direction information; Based on the driving trajectory data, identifying whether the electric bicycle has an abnormal speed event, an abnormal direction event, and an abnormal trajectory event; When at least one of an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is identified, the abnormal vehicle video data of the vehicle before and after 30 seconds of the driving recorder is intercepted; Abnormal speed events, abnormal direction events, abnormal trajectory events and abnormal vehicle video data are sent to the target device.

3. A data processing method according to claim 2, characterized in that: The data processing method comprises the following vehicle dumping identification steps: When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained; Detecting a third target object in the image data and generating a corresponding third target detection frame; Compare the image data with the third target detection frame with the template image, and identify whether the electric bicycle has fallen over according to the position relationship of the third target detection frame in the template image; When an electric bicycle is identified to have fallen over, a vehicle falling over alarm message is generated; Send vehicle tipping alarm information and image data to the target device.

4. A data processing method according to claim 2, characterized in that: The data processing method comprises the following vehicle dumping identification steps: When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained; detecting a third target object in the image data; identifying an outer contour of a third target object; Calculating a target area of ​​the third target object based on an outer contour of the third target object; Calculating a percentage value of the target area to the area of ​​the image data; When the percentage value is greater than a preset threshold, a vehicle tipping alarm message is generated; Send vehicle tipping alarm information and image data to the target device.

5. A data processing system, characterized in that: The data processing system is implemented based on a driving recorder, which is installed on an electric bicycle. The data processing system includes: An acquisition module, which is used to acquire video data of a driving recorder when recognizing that an electric bicycle is running; A detection module, which is used to detect a first target object and a second target object in the video data, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a non-motor vehicle, and the second target object is a motor vehicle; A tracking module, which is used to track the first target object and the second target object respectively using a multi-target tracking algorithm; A trajectory prediction module, which is used to use a trajectory prediction model to predict the trajectories of the first target object and the second target object respectively, to obtain a first predicted trajectory and a second predicted trajectory; A calculation module, which is used to calculate whether there is an accident conflict point between the first predicted trajectory and the second predicted trajectory; A generation module, which is used to generate road accident warning information when there is an accident conflict point; The sending module sends the accident warning information and video data to the target device.

6. A data processing system according to claim 5, characterized in that: The data processing system is also used to perform the following vehicle abnormality warning steps: Acquire the driving trajectory data of the electric bicycle in real time, wherein the driving trajectory data includes position information, speed information and direction information; Based on the driving trajectory data, identifying whether the electric bicycle has an abnormal speed event, an abnormal direction event, and an abnormal trajectory event; When at least one of an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is identified, the abnormal vehicle video data of the vehicle before and after 30 seconds of the driving recorder is intercepted; Abnormal speed events, abnormal direction events, abnormal trajectory events and abnormal vehicle video data are sent to the target device.

7. A data processing system according to claim 5, characterized in that: The data processing system is also used to perform the following vehicle dumping identification steps: When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained; Detecting a third target object in the image data and generating a corresponding third target detection frame; Compare the image data with the third target detection frame with the template image, and identify whether the electric bicycle has fallen over according to the position relationship of the third target detection frame in the template image; When an electric bicycle is identified to have fallen over, a vehicle falling over alarm message is generated; Send vehicle tipping alarm information and image data to the target device.

8. A data processing system according to claim 5, characterized in that: The data processing system is also used to perform the following vehicle dumping identification steps: When at least one of an accident warning message, an abnormal speed event, an abnormal direction event, and an abnormal trajectory event is generated, image data captured by the driving recorder 30 seconds later is obtained; detecting a third target object in the image data; identifying an outer contour of a third target object; Calculating a target area of ​​the third target object based on an outer contour of the third target object; Calculating a percentage value of the target area to the area of ​​the image data; When the percentage value is greater than a preset threshold, a vehicle tipping alarm message is generated; Send vehicle tipping alarm information and image data to the target device.

9. A driving recorder, characterized in that: The driving recorder is provided with a data processing system according to any one of claims 5 to 8.

Images