Accident injury severity classification method, electronic device, driving device, and storage medium

Abstract

The present application relates to the technical field of in-vehicle emergency rescue, and specifically provides an accident injury severity classification method, an electronic device, a driving device, and a storage medium, aiming to solve the technical problem that actual injury severities of occupants cannot be determined in time when a collision occurs in a driving device. For this purpose, the method of the present application comprises: in response to a collision occurring in a driving device, acquiring traveling data of the driving device; inputting the traveling data into a safety belt analysis model to obtain a safety belt analysis result; acquiring accident injury severity classification characteristics on the basis of the traveling data; and inputting the safety belt analysis result and the accident injury severity classification characteristics into an injury severity classification model to obtain a predicted injury severity classification result. By means of the method, when an accident occurs, safety belt wearing conditions of occupants can be analyzed in time, and the injury severity of each occupant is accurately predicted on the basis of the safety belt analysis result and the traveling data, thereby obtaining a predicted injury severity classification result.

Description

Accident injury classification method, electronic device, driving device and storage medium

[0001] This application claims priority to Chinese patent application No. 202410382633.7 filed on March 29, 2024, entitled “Accident Injury Classification Method, Electronic Device, Driving Device and Storage Medium”. The entire contents of the above Chinese patent application are incorporated into this application by reference. Technical Field

[0002] The present application relates to the field of vehicle-mounted rescue technology, and in particular to an accident injury classification method, electronic equipment, driving equipment, and storage medium. Background Art

[0003] At present, driving equipment is generally equipped with an emergency rescue call system, which can automatically make a rescue call after detecting a collision of the driving equipment and send the precise location of the driving equipment to the rescue center.

[0004] However, in actual situations, the severity of accidents varies. Sometimes the driver or passengers may be seriously injured or unconscious and unable to respond to rescue calls. At this time, it is impossible to judge the actual injury of the occupants in time.

[0005] Accordingly, this field requires a new technical solution to solve the above problems. Summary of the Invention

[0006] In order to overcome the above-mentioned defects, the present application is proposed to provide an accident injury classification method, electronic equipment, driving equipment and storage medium that solves or at least partially solves the technical problem that the actual injury status of the occupants cannot be determined in time after a collision of the driving equipment.

[0007] In a first aspect, a method for grading accident injuries is provided, the method comprising:

[0008] In response to a collision of the driving device, obtaining driving data of the driving device;

[0009] Inputting the driving data into a seat belt analysis model to obtain a seat belt analysis result;

[0010] Obtaining accident injury classification characteristics based on the driving data;

[0011] The seat belt analysis result and the accident injury classification feature are input into an injury classification model to obtain an injury classification result.

[0012] In one technical solution of the above-mentioned accident injury classification method, the driving data includes periodic operating status data, and the periodic operating status data includes trip information; inputting the driving data into a seat belt analysis model to obtain a seat belt analysis result includes:

[0013] Obtaining a seat occupancy signal and a seat belt signal from the trip information;

[0014] Inputting the seat occupancy signal and the seat belt signal into a seat belt analysis model to obtain the seat belt analysis result;

[0015] The seat belt analysis result includes whether each passenger in the cockpit of the driving device has engaged in seat belt cheating behavior.

[0016] In one technical solution of the above accident injury classification method, inputting the seat occupancy signal and the seat belt signal into a seat belt analysis model to obtain the seat belt analysis result includes:

[0017] Obtaining the signal status and signal start time of the seat occupancy signal and the seat belt signal;

[0018] Based on the signal state and the signal start time, it is determined whether each occupant in the cockpit of the driving device has a seat belt cheating behavior.

[0019] In one technical solution of the above-mentioned accident injury classification method, the driving data further includes triggered collision data; and obtaining injury classification features based on the driving data includes:

[0020] Analyzing the triggered collision data to obtain the injury classification characteristics;

[0021] The injury classification feature includes at least one of speed change information, occupant information, airbag status information, collision information, vehicle body posture information and alarm information.

[0022] In one technical solution of the above-mentioned accident injury classification method, the method further includes training the injury classification model based on at least the following steps:

[0023] Obtaining seat belt analysis results and injury classification features from crash test data and / or historical crash data as a training sample set;

[0024] Training the injury classification model based on the training sample set, and establishing a correlation between the seat belt analysis result and the injury classification feature and the injury classification level;

[0025] When the injury classification model converges to a preset error, the training of the injury classification model is completed.

[0026] In one technical solution of the above-mentioned accident injury classification method, inputting the seat belt analysis result and the injury classification feature into an injury classification model to obtain a predicted injury classification result includes:

[0027] The predicted injury grade result is obtained based on the correlation between the seat belt analysis result and the injury grade feature and the injury grade.

[0028] In one technical solution of the above-mentioned accident injury classification method, the method further includes:

[0029] Acquiring cockpit image data of the driving device;

[0030] The seat belt analysis result is verified based on the cockpit image data.

[0031] In one technical solution of the above-mentioned accident injury classification method, the injury classification result includes the injury level of each occupant in the cockpit of the driving equipment, and the injury levels include no injury, minor injury, moderate injury, serious injury and death.

[0032] In one technical solution of the above-mentioned accident injury classification method, the method further includes:

[0033] In response to a collision of the driving device, sending an emergency rescue signal;

[0034] and / or,

[0035] The predicted injury classification result is sent to the rescue service platform.

[0036] In one technical solution of the above-mentioned accident injury classification method, the method further includes:

[0037] Obtain the actual injury level information of each occupant in the cockpit of the driving equipment;

[0038] Feedback training is performed on the injury grading model based on the actual injury level information.

[0039] In a second aspect, an electronic device is provided, comprising:

[0040] at least one processor;

[0041] and, a memory communicatively coupled to the at least one processor;

[0042] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the accident injury classification method described in any one of the technical solutions of the above-mentioned accident injury classification method is implemented.

[0043] In a third aspect, a driving device is provided, which includes a driving device body and the electronic device described in the technical solution of the above electronic device.

[0044] In a fourth aspect, a computer-readable storage medium is provided, in which a plurality of program codes are stored. The program codes are suitable for being loaded and run by a processor to execute the accident injury classification method described in any one of the technical solutions of the above-mentioned accident injury classification method.

[0045] The above one or more technical solutions of this application have at least one or more of the following beneficial effects:

[0046] In implementing the technical solution of this application, in response to a collision involving a driving device, driving data from the driving device is acquired, the driving data is input into a seatbelt analysis model to obtain a seatbelt analysis result, and accident injury classification features are obtained based on the driving data. The seatbelt analysis results and the accident injury classification features are input into the injury classification model to obtain an injury classification result. Through the above-described implementation, the seatbelt wearing status of occupants can be promptly analyzed in the event of an accident, and the injury status of each occupant can be accurately predicted based on the seatbelt analysis results and driving data, resulting in a predicted injury classification result.

[0047] Furthermore, the predicted injury classification results can be sent to the rescue service platform to guide the emergency rescue team to allocate resources, provide the most timely and appropriate rescue measures, and improve the emergency response speed and rescue efficiency after the accident. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The disclosure of this application will become more easily understood with reference to the accompanying drawings. Those skilled in the art will readily appreciate that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Among them:

[0049] FIG1 is a flow chart showing the main steps of a method for grading accident injuries according to an embodiment of the present application;

[0050] FIG2 is a flow chart showing the main steps of a method for grading accident injuries according to another embodiment of the present application;

[0051] FIG3 is a flow chart showing the main steps of a method for training an injury classification model according to an embodiment of the present application;

[0052] FIG4 is a flow chart showing the main steps of a method for grading accident injuries according to another embodiment of the present application;

[0053] FIG5 is a schematic diagram of the main process of a method for grading accident injuries according to an embodiment of the present application;

[0054] FIG6 is a main structural block diagram of an accident injury classification device according to an embodiment of the present application;

[0055] FIG7 is a schematic diagram of the main structure of an electronic device according to an embodiment of the present application.

[0056] List of reference numerals:

[0057] 601: First acquisition module; 602: Analysis module; 603: Second acquisition module; 604: Injury classification module; 701: Processor; 702: Memory. DETAILED DESCRIPTION

[0058] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the scope of protection of the present application.

[0059] In the description of this application, "module" and "processor" may include hardware, software, or a combination of both. A module may include hardware circuitry, various suitable sensors, communication ports, and memory. It may also include software components, such as program code, or a combination of software and hardware. A processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. A processor has data and / or signal processing capabilities. A processor may be implemented in software, hardware, or a combination of both. Non-transitory computer-readable storage media include any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" refers to all possible combinations of A and B, such as only A, only B, or both A and B. The terms "at least one of A or B" or "at least one of A and B" have similar meanings to "A and / or B" and may include only A, only B, or both A and B. The singular forms "a" and "the" may also include the plural forms.

[0060] Here we first explain some terms involved in this application.

[0061] EDR (Event Data Recorder): An event data recording system used to record key vehicle operating data (including speed, ABS status, seat belt status, etc.) in three stages: before, during, and after a collision.

[0062] Delta V: Delta V is the instantaneous change in the vehicle's speed when a collision occurs.

[0063] OLC (Occupant Load Criterion) is an indicator for evaluating vehicle deceleration. It is also based on the average occupant deceleration calculated by assuming that the occupants are moving forward under the given vehicle deceleration waveform. It is used to evaluate the load on the occupants caused by vehicle deceleration.

[0064] As described in the background art, driving devices are generally equipped with an emergency rescue call system, which can automatically dial a rescue phone after detecting a collision of the driving device and send the precise location of the driving device to a rescue center.

[0065] However, in actual situations, the severity of accidents varies. Sometimes the driver or passengers may be seriously injured or unconscious and unable to respond to rescue calls. At this time, it is impossible to judge the actual injury of the occupants in time.

[0066] In order to solve the above problems, the present application provides an accident injury classification method, electronic equipment, driving equipment and storage medium.

[0067] Referring to FIG1 , FIG1 is a flow chart showing the main steps of a method for grading injury severity according to an embodiment of the present application. As shown in FIG1 , the method for grading injury severity according to the embodiment of the present application mainly includes the following steps S101 to S104.

[0068] Step S101: In response to a collision of a driving device, obtaining driving data of the driving device;

[0069] Step S102: inputting the driving data into the seat belt analysis model to obtain the seat belt analysis results;

[0070] Step S103: Obtaining accident injury classification characteristics based on driving data;

[0071] Step S104: inputting the seat belt analysis results and the accident injury classification characteristics into the injury classification model to obtain a predicted injury classification result.

[0072] Based on the above steps S101 to S104, the seat belt wearing status of the occupants can be analyzed in a timely manner when an accident occurs, and the injury status of each occupant can be accurately predicted based on the seat belt wearing status and driving data to obtain a predicted injury classification result.

[0073] The above steps S101 to S104 are further explained below.

[0074] The accident injury classification method in the embodiment of the present application can be applied to a secure cloud platform, which can acquire data and perform cloud computing on the data.

[0075] In some implementations of the above step S101, in response to a collision of the driving device, driving data of the driving device may be acquired, wherein the driving data includes periodic operating state data and triggered collision data.

[0076] Specifically, the vehicle data loopback mechanism supports two types of data collection, namely periodic operating status data and triggered collision data.

[0077] The periodic operating status data is the operating status data uploaded by the driving device at a set time interval. The collection frequency of the periodic operating status data is low and the data volume is relatively small. It can include some data related to the driving device collision.

[0078] When a driving device collides, the collected periodic operation status data will be uploaded to the database. Furthermore, the security cloud platform can read the periodic operation status data in the database.

[0079] The triggered collision data is raw CAN frequency data, which has a high acquisition frequency, high data accuracy, and a large data volume, and can include a variety of data related to the collision of the driving device.

[0080] When a collision occurs in a driving device, the collected raw CAN frequency data files are packaged and transmitted to the database. Furthermore, the security cloud platform can obtain the triggered collision data in the database through the TSP (Telematics Service Provider) interface.

[0081] The triggered collision data and the periodic operation status data may be stored in the same database or in different databases, which is not limited here.

[0082] Furthermore, the periodic operation status data may include trip information, such as trip ID, seat occupancy signal, seat belt signal, and other information.

[0083] Triggered collision data can include the following data:

[0084] Vehicle type: including vehicle dimensions such as length, width, height and weight;

[0085] Speed ​​change information: including horizontal and vertical acceleration, OLC and Delta V value, etc. Among them, the horizontal and vertical acceleration and Delta V value can be read from the EDR, and OLC can be calculated from the horizontal and vertical acceleration read from the EDR;

[0086] Tire pressure: can be obtained through the tire pressure monitoring system;

[0087] The occupancy status of each seat in the cabin and the seat belt status can be obtained through the internal sensors of the driving equipment;

[0088] Airbag deployment signal: including the number of airbags deployed, deployment status of the main airbag, side curtain airbag and side airbag, etc., which can be read from the EDR;

[0089] Collision information: including collision surface judgment signal, time from collision occurrence to vehicle stabilization, and collision count, which can be obtained through collision sensors and / or read from EDR;

[0090] Vehicle posture signals: including yaw, pitch, and roll, which can be obtained through sensors;

[0091] Passenger data: This includes the age and gender data of each passenger. Specifically, passenger images can be obtained through the vehicle-side Occupancy Monitoring System (OMS) and processed on the vehicle side, such as by analyzing passenger images using computer vision algorithms and machine learning algorithms to obtain the age and gender data of each passenger. The passenger's gender data can be represented by binary data, such as "0" for male and "1" for female. The age data can be a detailed age or an age range, such as children, youth, middle-aged and elderly. It can also be used to determine whether the passenger meets a preset age threshold, such as whether he or she is over 55 years old.

[0092] Various alarm information: including brake system alarm information, engine alarm information and airbag alarm information, etc.

[0093] Among them, EDR (Event Data Recorder) is an event data recording system used to record the key operating data of the car in three stages: before, during and after a vehicle collision.

[0094] OLC (Occupant Load Criterion) is an indicator for evaluating vehicle deceleration. Given a given vehicle deceleration waveform, it assumes the occupants are moving purely forward, and calculates the average occupant deceleration. This is used to evaluate the load exerted by vehicle deceleration on the occupants.

[0095] Delta V is the change in velocity or relative velocity, which refers to the instantaneous change in vehicle velocity when a collision occurs.

[0096] The above is an explanation of the periodic operation status data and the triggered collision data.

[0097] It should be noted that the above examples of triggered collision data, periodic operation status data and corresponding acquisition methods are only illustrative. In actual applications, those skilled in the art can collect them according to specific needs, and no limitation is made here.

[0098] The dual data collection mechanism not only improves the integrity and reliability of driving data, but also provides strong data support for subsequent accident analysis.

[0099] The above is the description of step S101.

[0100] Furthermore, in some embodiments, the accident injury classification method provided by the present application further includes: sending an emergency rescue signal in response to a collision of the driving device.

[0101] Specifically, when a collision occurs with a driving device, an emergency rescue signal can be automatically sent to a rescue service platform through the emergency rescue call system, and the location information of the driving device can also be sent at the same time.

[0102] The following further describes step S102.

[0103] In some implementations of the above step S102 , a seat occupancy signal and a seat belt signal in the travel information may be obtained, and the seat occupancy signal and the seat belt signal may be input into a seat belt analysis model to obtain a seat belt analysis result.

[0104] Among them, the seat belt analysis results include whether each occupant in the cockpit of the driving equipment has engaged in seat belt cheating behavior.

[0105] According to a big data survey, 3% of users do not like to wear seat belts during their trips. However, in order to evade traffic monitoring or prevent alarms, they will engage in some seat belt cheating behaviors, such as using special seat belt buckles or other items to deceive the seat belt detection system, partially fastening the seat belt but not tightening it to fit the body, installing fake fasteners on the seat belt or using materials such as transparent tape to pretend that the seat belt is fastened, using clothing or other items to cover the seat belt, etc., so that the seat belt detection system shows that the seat belt is being worn when it is not actually being worn.

[0106] Not wearing a seat belt significantly increases the risk of serious injury or death to passengers in the event of a vehicle collision. When passengers engage in the above-mentioned seat belt cheating behaviors, the seat belt signals obtained will be incorrect, thereby affecting the judgment of the accident.

[0107] Therefore, the seat occupancy signal and seat belt signal in the travel data can be input into the trained seat belt analysis model to obtain the seat belt analysis results.

[0108] Among them, the seat belt analysis model can perform logical analysis on the input seat occupancy signal and seat belt signal.

[0109] In some embodiments, the seat belt analysis model can obtain the signal status and signal start time of the seat occupancy signal and the seat belt signal, and determine whether each occupant in the driving device cabin has engaged in seat belt cheating based on the signal status and signal start time.

[0110] For example, after a trip begins, the seat belt signal is first detected as being worn, followed by the seat occupied signal, and the seat belt signal's start time is earlier than the seat occupied signal's start time, and the signal state does not change during the trip. This indicates that the occupant inserted the seat belt buckle before getting into their seat, and this occupant is engaging in seat belt cheating. If, after a trip begins, the seat occupied signal is first detected as being occupied, followed by the seat belt signal's state being worn, and the seat occupied signal's start time is earlier than the seat belt signal's start time, and the signal state does not change during the trip, this indicates that the occupant inserted the seat belt buckle after getting into their seat, and this occupant is not engaging in seat belt cheating.

[0111] Furthermore, in order to improve the accuracy of seat belt analysis, an in-cabin camera can be introduced to perform visual algorithms to verify the seat belt analysis results.

[0112] In some embodiments, referring to FIG2 , FIG2 is a flow chart of the main steps of the accident injury classification method according to another embodiment of the present application. As shown in FIG2 , the method mainly includes the following steps S201 to S202:

[0113] Step S201: Acquire cockpit image data of a driving device;

[0114] Specifically, the cockpit image of the driving device can be first obtained through the vehicle-side Occupancy Monitoring System (OMS), and then the cockpit image can be processed on the vehicle side to obtain cockpit image data, and finally the cockpit image data can be uploaded to the secure cloud platform.

[0115] The vehicle side can use image processing technologies, such as computer vision and machine learning algorithms, to analyze the seatbelt status in the cabin image and identify whether the seatbelt is being worn correctly. The processed cabin image data can be represented by a non-sensitive binary value, such as "0" for not wearing a seatbelt and "1" for wearing a seatbelt.

[0116] Step S202: Verify the seat belt analysis result based on the cockpit image data.

[0117] Verifying seatbelt analysis results using cockpit image data can improve the accuracy of seatbelt cheating analysis and further accurately predict the injury status of each occupant.

[0118] Among them, the cockpit image data can be collected and uploaded when the driving equipment collides, or it can be uploaded as triggered collision data or periodic operating status data, which is not limited here.

[0119] The above is a further description of step S102 , and the following further describes step S103 .

[0120] In some implementations of the above step S103, injury classification features may be obtained based on driving data.

[0121] Specifically, the triggered collision data can be parsed to obtain injury classification features. That is, the original CAN frequency data files packaged in the database are parsed to obtain injury classification features that need to be input into the injury classification model.

[0122] In some embodiments, after analyzing the triggered collision data, steps such as data cleaning, missing value processing, outlier processing, feature selection, feature encoding, normalization or standardization may be performed to obtain standard injury classification features.

[0123] Among them, the injury classification features may include one or more of speed change information, occupant information, airbag status information, collision information, vehicle posture information, alarm information, vehicle model, tire pressure, occupancy status of each seat in the cabin / seat belt status, etc., which are not limited here.

[0124] Furthermore, speed change information may include lateral and longitudinal acceleration, OLC and Delta V values; occupant information may include age information and gender information of each occupant; airbag status information may include the number of airbags deployed, the deployment status of the main airbag, side air curtain and side airbags, etc.; collision information may include collision surface information, the time from the occurrence of the collision to the entire vehicle being in a stable state and the number of collisions, etc.; vehicle posture information may include yaw, pitch and roll, etc.; alarm information may include braking system alarm information, engine alarm information and airbag alarm information, etc.; vehicle model may include the length, width, height and weight of the vehicle, etc.; tire pressure includes the pressure of each tire; seat occupancy status / seat belt status may include the signal status of the seat occupancy signal / seat belt signal, signal start time, etc.

[0125] It should be pointed out that the above examples of injury classification characteristics are only illustrative. In actual applications, those skilled in the art can classify injury characteristics according to specific needs, and no limitation is made here.

[0126] The above is a further description of step S103 , and the following further describes step S104 .

[0127] In some implementations of the above step S104, the injury classification model is pre-trained, and the injury classification model can obtain a predicted injury classification result based on the input seat belt analysis result and injury classification features.

[0128] Refer to FIG3 , which is a flowchart of the main steps of the injury classification model training method according to an embodiment of the present application. As shown in FIG3 , the injury classification model can be trained through the following steps S301 to S303:

[0129] Step S301: Acquire seat belt analysis results and injury classification features from collision test data and / or historical collision data as a training sample set;

[0130] Specifically, collision test data and / or historical collision data marked with injury severity levels may be obtained. The collision test data is data obtained from collision tests conducted under laboratory conditions, and the historical collision data is data from real accident cases.

[0131] Furthermore, the collision test data and / or historical collision data may be preprocessed, such as data cleaning, normalization, enhancement, etc., to obtain a training sample set.

[0132] Step S302: training the injury classification model based on the training sample set, and establishing a correlation between the seat belt analysis results and the injury classification features and the injury classification level;

[0133] Among them, the injury classification model can be constructed based on the existing logistic regression model.

[0134] When the injury classification model is trained based on the training sample set, the collision test data and / or historical collision data can be analyzed and counted to establish a correlation between the seat belt analysis results and the injury classification features and the injury grade.

[0135] Step S303: When the injury classification model converges to a preset error, the injury classification model training is completed.

[0136] Specifically, the injury classification model will output the predicted injury classification results based on the seat belt analysis results and injury classification characteristics in the training samples, and will continuously train the model based on the degree of similarity between the model-predicted injury classification results and the occupants' actual injury levels as the optimization target, update the model parameters, and adjust and optimize the model until the model converges to the preset error, completing the training.

[0137] Furthermore, the seat belt analysis results and accident injury classification characteristics can be input into the injury classification model, and the predicted injury classification results can be obtained based on the correlation between the seat belt analysis results and the injury classification characteristics and the injury classification level.

[0138] The predicted injury classification result includes the injury classification of each occupant in the cockpit of the driving device, and specifically, the injury classification probability of each occupant. The injury classification may include no injury, minor injury, moderate injury, severe injury, and death.

[0139] For example, within the cockpit of a driving device, there are three passengers each in the driver's seat, front passenger seat, and second row on the left. After a collision, the seatbelt analysis results, based on seat occupancy signals and seatbelt signals, indicate that the occupants in the driver's and front passenger seats did not engage in seatbelt cheating, while the occupant in the second row on the left did. The accident injury classification features derived from triggered collision data include collision information and airbag status information. The collision surface is the left side, and four airbags were deployed: the driver's side airbag, the right curtain airbag, the left curtain airbag, and the far center airbag.

[0140] After the above-mentioned seat belt analysis results and accident injury classification characteristics are input into the injury classification model, the predicted injury classification results are shown in Table 1 below based on the correlation between the seat belt analysis results and the injury classification characteristics and the injury classification.

[0141] Table 1

[0142] Furthermore, in some embodiments, after obtaining the predicted injury classification result, the predicted injury classification result can be sent to a rescue service platform.

[0143] Specifically, the security cloud platform can send the predicted injury classification results directly to the rescue service platform; the security cloud platform can also send the predicted injury classification results to the emergency rescue call system, and the emergency rescue call system will send the predicted injury classification results to the rescue service platform after receiving them. There is no limitation here.

[0144] Through the above implementation methods, the emergency rescue team can be guided to allocate resources, provide the most timely and appropriate rescue measures, and improve the emergency response speed and rescue efficiency after an accident occurs.

[0145] Further, in some embodiments, referring to FIG4 , FIG4 is a flow chart of the main steps of the accident injury classification method according to another embodiment of the present application. As shown in FIG4 , the method mainly includes the following steps S401 to S402:

[0146] Step S401: Acquire the actual injury level information of each occupant in the cockpit of the driving equipment;

[0147] Specifically, after the rescue personnel arrive at the scene, they can understand the actual injury of each occupant in the accident, grade and label the actual injury of each occupant according to the injury grading standards, and obtain the actual injury level information of each occupant.

[0148] Furthermore, the actual injury level information of each occupant can be uploaded to the security cloud platform, or uploaded to the emergency rescue call system, and sent to the security cloud platform by the emergency rescue call system.

[0149] Step S402: performing feedback training on the injury grading model based on actual injury grade information.

[0150] Specifically, the actual injury grade information can be input into the injury grade model, and the model can be iteratively updated based on the difference between the predicted injury grade result and the actual injury grade information, and the model parameters can be adjusted to improve the accuracy and robustness of the model.

[0151] The above examples of feedback training are only illustrative. In actual applications, those skilled in the art can perform corresponding feedback training on the injury grading model according to specific scenarios and needs, which is not limited here.

[0152] The above is an explanation of the accident injury classification method provided in this application.

[0153] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of the present application, different steps do not have to be performed in such an order. They can be performed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present application.

[0154] For example, the above-mentioned steps S102 and S103 can be executed simultaneously, that is, the driving data is input into the seat belt analysis model to obtain the seat belt analysis results, and the accident injury classification characteristics are obtained based on the driving data; or step S102 can be executed first, and then step S103 is executed, that is, the driving data is first input into the seat belt analysis model to obtain the seat belt analysis results, and then the accident injury classification characteristics are obtained based on the driving data; or step S103 can be executed first, and then step S102 is executed, that is, the accident injury classification characteristics are first obtained based on the driving data, and then the driving data is input into the seat belt analysis model to obtain the seat belt analysis results. This is not limited here.

[0155] Furthermore, in some implementations, referring to FIG5 , FIG5 is a schematic diagram of the main flow of an accident injury classification method according to an embodiment of the present application.

[0156] As shown in Figure 5, after a collision occurs with the driving device, the triggered collision data and the periodic operation status data are uploaded to different databases respectively;

[0157] The safety cloud platform obtains triggered collision data from the database through the TSP interface and analyzes the data to obtain accident injury classification characteristics. It also reads the trip information from the database and inputs it into the seatbelt analysis model. Furthermore, the seatbelt analysis results and accident injury classification characteristics are input into the injury classification model to obtain a predicted injury classification result. This predicted injury classification result is then sent to the emergency rescue call system.

[0158] The emergency rescue call system sends the received predicted injury level results to the rescue service platform, and receives the actual injury level information sent by the rescue service platform; further, the actual injury level information is sent to the security cloud platform;

[0159] The security cloud platform provides feedback training for the injury classification model based on the actual injury level information received.

[0160] For the convenience and brevity of description, the specific working process and related instructions of the accident injury classification method shown in FIG5 can refer to the contents described in the embodiment of the above steps S101 to S104, which will not be repeated here.

[0161] Through the above implementation, when an accident occurs, the seat belt wearing status of the occupants can be analyzed in a timely manner, and the injury status of each occupant can be accurately predicted based on the seat belt analysis results and driving data to obtain a predicted injury classification result.

[0162] Furthermore, the predicted injury classification results can be sent to the rescue service platform to guide the emergency rescue team to allocate resources, provide the most timely and appropriate rescue measures, and improve the emergency response speed and rescue efficiency after the accident.

[0163] Furthermore, the present application also provides an accident injury grading device.

[0164] Refer to Figure 6, which is a main structural block diagram of an accident injury classification device according to an embodiment of the present application. As shown in Figure 6, the accident injury classification device in the embodiment of the present application mainly includes a first acquisition module 601, an analysis module 602, a second acquisition module 603 and an injury classification module 604. In some embodiments, the first acquisition module 601 can be configured to obtain driving data of the driving device in response to a collision of the driving device. The analysis module 602 can be configured to input the driving data into a seat belt analysis model to obtain a seat belt analysis result. The second acquisition module 603 can be configured to obtain accident injury classification features based on the driving data. The injury classification module 604 can be configured to input the seat belt analysis results and the accident injury classification features into the injury classification model to obtain a predicted injury classification result.

[0165] In one implementation, the description of specific implementation functions can refer to steps S101 to S104.

[0166] Furthermore, it should be understood that since the configuration of each module is merely for the purpose of illustrating the functional units of the apparatus of the present application, the physical devices corresponding to these modules may be the processor itself, or a portion of the software in the processor, a portion of the hardware, or a combination of software and hardware. Therefore, the number of modules in the figure is merely illustrative.

[0167] Those skilled in the art will appreciate that the various modules in the device can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principles of this application. Therefore, the technical solutions after splitting or merging will fall within the scope of protection of this application.

[0168] The above-mentioned accident injury grading device can be used to execute the embodiment of the accident injury grading method shown in Figure 1. The technical principles, technical problems solved and technical effects produced by the two are similar. Technical personnel in this technical field can clearly understand that for the convenience and conciseness of description, the specific working process and related instructions of the accident injury grading device can refer to the contents described in the embodiment of the accident injury grading method, and will not be repeated here.

[0169] It will be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above-mentioned various method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0170] Furthermore, the present application also provides an electronic device. Refer to Figure 7, which is a schematic diagram of the main structure of an electronic device according to an embodiment of the present application. As shown in Figure 7, the electronic device in the embodiment of the present application mainly includes a processor 701 and a memory 702. The memory 702 can be configured to store a program for executing the accident injury classification method of the above-mentioned method embodiment, and the processor 701 can be configured to execute the program in the memory 702, which includes but is not limited to the program for executing the accident injury classification method of the above-mentioned method embodiment. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application.

[0171] In some possible implementations of the present application, the electronic device may include multiple processors 701 and multiple memories 702. The program for executing the accident injury grading method of the above-mentioned method embodiment can be divided into multiple subroutines, and each subroutine can be loaded and run by the processor 701 to execute different steps of the accident injury grading method of the above-mentioned method embodiment. Specifically, each subroutine can be stored in a different memory 702, and each processor 701 can be configured to execute the program in one or more memories 702 to jointly implement the accident injury grading method of the above-mentioned method embodiment, that is, each processor 701 executes different steps of the accident injury grading method of the above-mentioned method embodiment to jointly implement the accident injury grading method of the above-mentioned method embodiment.

[0172] The multiple processors 701 may be processors deployed on the same device. For example, the electronic device may be a high-performance device composed of multiple processors, and the multiple processors 701 may be processors configured on the high-performance device. Furthermore, the multiple processors 701 may be processors deployed on different devices. For example, the electronic device may be a server cluster, and the multiple processors 701 may be processors on different servers in the server cluster.

[0173] Furthermore, the present application also provides a driving device. In an embodiment of a driving device according to the present application, the driving device may include a driving device body and the electronic device described in the above electronic device embodiment.

[0174] Furthermore, the present application also provides a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present application, the computer-readable storage medium can be configured to store a program for executing the accident injury classification method of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned accident injury classification method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium can be a memory device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.

[0175] It should be noted that the relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, follows the principles of legality, legitimacy and necessity, and is based on the reasonable purposes of business scenarios to process personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as personal information obtained with the user's authorization.

[0176] The user personal information processed by this application will vary depending on the specific product / service scenario and must be based on the specific scenario in which the user uses the product / service. This may involve the user's account information, device information, driving information, vehicle information, or other related information. This application will treat the user's personal information and its processing with a high degree of diligence.

[0177] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that comply with industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0178] Thus far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A method for grading accident injuries, characterized in that: The method comprises: In response to a collision of the driving device, obtaining driving data of the driving device; Inputting the driving data into a seat belt analysis model to obtain a seat belt analysis result; obtaining injury classification characteristics based on the driving data; The seat belt analysis result and the injury classification feature are input into an injury classification model to obtain a predicted injury classification result.

2. The accident injury classification method according to claim 1, characterized in that: The driving data includes periodic operating status data, and the periodic operating status data includes travel information; inputting the driving data into a seat belt analysis model to obtain a seat belt analysis result includes: Obtaining a seat occupancy signal and a seat belt signal from the trip information; Inputting the seat occupancy signal and the seat belt signal into a seat belt analysis model to obtain the seat belt analysis result; The seat belt analysis result includes whether each passenger in the cockpit of the driving device has engaged in seat belt cheating behavior.

3. The accident injury classification method according to claim 2, characterized in that: Inputting the seat occupancy signal and the seat belt signal into a seat belt analysis model to obtain the seat belt analysis result includes: Obtaining the signal status and signal start time of the seat occupancy signal and the seat belt signal; Based on the signal state and the signal start time, it is determined whether each occupant in the cockpit of the driving device has a seat belt cheating behavior.

4. The accident injury classification method according to claim 1, characterized in that: The driving data also includes triggered collision data; and obtaining injury classification features based on the driving data includes: Analyzing the triggered collision data to obtain the injury classification characteristics; The injury classification feature includes at least one of speed change information, occupant information, airbag status information, collision information, vehicle body posture information and alarm information.

5. The accident injury classification method according to claim 1, characterized in that: The method further includes training the injury classification model based on at least the following steps: Obtaining seat belt analysis results and injury classification features from crash test data and / or historical crash data as a training sample set; Training the injury classification model based on the training sample set, and establishing a correlation between the seat belt analysis result and the injury classification feature and the injury classification level; When the injury classification model converges to a preset error, the training of the injury classification model is completed.

6. The accident injury classification method according to claim 5, characterized in that: Inputting the seat belt analysis result and the injury classification feature into an injury classification model to obtain a predicted injury classification result includes: The predicted injury grade result is obtained based on the correlation between the seat belt analysis result and the injury grade feature and the injury grade.

7. The accident injury classification method according to claim 2, characterized in that: The method further comprises: Acquiring cockpit image data of the driving device; The seat belt analysis result is verified based on the cockpit image data.

8. The accident injury classification method according to claim 1, characterized in that: The predicted injury classification result includes the injury classification of each occupant in the cockpit of the driving equipment, and the injury classification includes no injury, slight injury, moderate injury, severe injury and death.

9. The accident injury classification method according to claim 1, characterized in that: The method further comprises: In response to a collision of the driving device, sending an emergency rescue signal; and / or, The predicted injury classification result is sent to the rescue service platform.

10. The accident injury classification method according to claim 1, characterized in that: The method further comprises: Obtain the actual injury level information of each occupant in the cockpit of the driving equipment; Feedback training is performed on the injury grading model based on the actual injury level information.

11. An electronic device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the accident injury classification method according to any one of claims 1 to 10 is implemented.

12. A driving device, characterized in that: The driving device includes a driving device body and the electronic device according to claim 11.

13. A computer-readable storage medium storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and run by a processor to execute the accident injury classification method according to any one of claims 1 to 10.

Images