Traffic accident identification method, device, system and program product

Abstract

The invention relates to the field of artificial intelligence, in particular to a traffic accident identification method, device and system and a program product. The method comprises the steps that a vehicle end obtains multi-modal data, traffic accidents are preliminarily screened according to the multi-modal data, and relevant information of suspected traffic accidents is obtained; the vehicle end sends the multi-modal data and the suspected traffic accident related information to the platform end; the platform end determines personalized driving habits of a driver according to the driver information of the vehicle end, and determines historical accident information of the position according to the position of the suspected traffic accident; and the platform end obtains an identification result of the suspected traffic accident according to the personalized driving habit of the driver, the historical accident information of the position and the multi-modal data. According to the method, the historical accident information of the position of the suspected traffic accident and the personalized driving habit of the driver can be combined to analyze and identify the multi-modal data, so that a more accurate traffic accident identification result is obtained, false alarms are reduced, and the driving efficiency of a motorcade is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to methods, devices, systems and program products for identifying traffic accidents. Background Art

[0002] Traffic accident identification is an important research direction in the field of intelligent transportation systems and autonomous driving. It plays a vital role in improving road safety and traffic management. With the increasing complexity of transportation networks and the continuous growth of traffic volume, the importance of intelligent detection and identification of traffic accidents continues to rise.

[0003] Traditional accident identification algorithms usually collect data from the device and analyze the data in real time to determine whether there is a traffic accident at the current moment, and generate an accident alarm based on the analysis results. Since the accuracy of traffic accidents analyzed by the device is not high, false alarms are prone to occur, which can easily affect the driving efficiency of the fleet. SUMMARY OF THE INVENTION Technical issues

[0004] In view of this, the embodiments of the present application provide a traffic accident identification method, device, system and program product to solve the problems in the prior art of low accuracy of traffic accidents, easy occurrence of false alarms, and easy impact on the driving efficiency of the fleet. Technical Solutions

[0005] A first aspect of an embodiment of the present application provides a traffic accident identification method, the method comprising:

[0006] The vehicle side acquires multimodal data, and performs preliminary screening of traffic accidents based on the multimodal data to acquire information related to suspected traffic accidents, wherein the multimodal data includes driver information of the vehicle side, and the information related to suspected traffic accidents includes the location of the suspected traffic accident;

[0007] The vehicle end sends the multimodal data and the suspected traffic accident related information to the platform end;

[0008] The platform determines the driver's personalized driving habits based on the driver information of the vehicle, and determines the historical accident information of the location based on the location of the suspected traffic accident;

[0009] The platform obtains the identification result of the suspected traffic accident according to the driver's personalized driving habits, the historical accident information of the location and the multimodal data.

[0010] In combination with the first aspect, in a first possible implementation manner of the first aspect, the platform obtains the identification result of the suspected traffic accident according to the personalized driving habits of the driver, the historical accident information of the location, and the multimodal data, including:

[0011] The platform generates an accident description text based on the information related to the suspected traffic accident, the driver's personalized driving habits, and the historical accident information of the location;

[0012] The platform inputs the accident description text into a large language model to obtain a recognition result of the suspected traffic accident.

[0013] In combination with the first aspect, in a second possible implementation manner of the first aspect, the vehicle side sends the multimodal data and the suspected traffic accident related information to the platform side, and the method further includes:

[0014] After the vehicle end obtains the information related to the suspected traffic accident, the vehicle end continues to obtain multimodal data for a predetermined time period;

[0015] The vehicle side sends the multimodal data used to determine the information related to the suspected traffic accident, the multimodal data that continues to be acquired, and the traffic accident information to the platform side.

[0016] In combination with the first aspect, in a third possible implementation manner of the first aspect, after the platform obtains the identification result of the suspected traffic accident according to the driver's personalized driving habits, the historical accident information of the location, and the multimodal data, the method further includes:

[0017] The platform determines the accident level based on the vehicle's six-axis acceleration and vehicle speed acquired by the sensor and the weight information corresponding to the terrain at the vehicle's current location;

[0018] The platform determines the priority of handling the vehicle accident according to the accident level.

[0019] In combination with the first aspect, in a fourth possible implementation manner of the first aspect, the multimodal data includes vehicle motion data, in-cockpit image data, and out-cockpit image data;

[0020] Perform preliminary screening of traffic accidents based on the multimodal data to obtain relevant information of suspected traffic accidents, including:

[0021] The vehicle end determines at least one of a sudden deceleration of the vehicle, a collision of the vehicle, and a rollover of the vehicle according to the vehicle motion data;

[0022] The vehicle end determines at least one of an abnormal driver movement event and a driver panic event based on the in-cockpit image data.

[0023] In combination with any one of the first aspect to the fourth possible implementation manner of the first aspect, in a fifth possible implementation manner of the first aspect, the platform side obtains the identification result of the suspected traffic accident according to the personalized driving habits of the driver, the historical accident information of the location, and the multimodal data, including:

[0024] Determining a first relevant feature associated with historical accident information, and determining a second relevant feature associated with personalized driving habits;

[0025] The weight of the first relevant feature is determined according to the frequency of historical accident information at the location, and the weight of the second relevant feature is determined according to the personalized driving habit of the driver.

[0026] In combination with the fifth possible implementation manner of the first aspect, in a sixth possible implementation manner of the first aspect, determining the weight of the first relevant feature according to the frequency of the historical accident information of the location includes:

[0027] According to the frequency of historical traffic accidents at the location, the weight of the first relevant feature is adjusted in a proportional relationship;

[0028] determining a weight of the second relevant feature according to the personalized driving habits of the driver;

[0029] According to the degree of the driver's personalized driving habit, the weight of the second related feature is adjusted in an inverse proportional correspondence.

[0030] A second aspect of an embodiment of the present application provides a traffic accident identification device, the device comprising:

[0031] A data acquisition unit, configured to acquire multimodal data from a vehicle, and perform preliminary screening of traffic accidents based on the multimodal data to acquire information related to suspected traffic accidents, wherein the multimodal data includes information about the driver of the vehicle, and the information related to suspected traffic accidents includes the location of the suspected traffic accident;

[0032] An information sending unit, configured to send the multimodal data and the suspected traffic accident related information to the platform end from the vehicle end;

[0033] An information determination unit, configured to determine the driver's personalized driving habits according to the driver information of the vehicle end by the platform end, and determine the historical accident information of the location according to the location of the suspected traffic accident;

[0034] The accident identification unit is used for obtaining the identification result of the suspected traffic accident by the platform according to the personalized driving habits of the driver, the historical accident information of the location and the multimodal data.

[0035] A third aspect of an embodiment of the present application provides a traffic accident identification system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the traffic accident identification system implements a method as described in any one of the first aspects.

[0036] A fourth aspect of the embodiments of the present application provides a computer program product, which, when executed on a computer, enables the computer to execute the method in the first aspect or its various implementations.

[0037] A fifth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in any one of the first aspects are implemented.

[0038] The sixth aspect of the embodiment of the present application provides a chip for implementing the methods in each implementation of the first aspect. Specifically, the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip executes the method in the first aspect or its implementation. Beneficial Effects

[0039] Compared with the prior art, the embodiments of the present application have the following beneficial effects: the embodiments of the present application obtain multimodal data through the vehicle end, preliminarily screen the multimodal data through the vehicle end, obtain information related to suspected traffic accidents, and send the information related to suspected traffic accidents to the platform end. The platform end determines the driver's personalized driving habits based on the driver information, and determines the historical accident information of the location based on the location of the suspected traffic accident. The platform end performs specific screening and judgment based on the driver's personalized driving habits and the historical accident information of the location of the suspected traffic accident to obtain the identification result of the suspected traffic accident. Compared with the identification method of the vehicle end, this method can combine the historical accident information of the location of the suspected traffic accident and the personalized driving habits of the driver to analyze and identify the multimodal data, and obtain more accurate traffic accident identification results, which is conducive to reducing false alarms and improving the driving efficiency of the fleet. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0041] Figure 1 This is a schematic diagram of an application scenario of a traffic accident identification method provided in an embodiment of the present application;

[0042] Figure 2 It is a schematic diagram of the implementation process of a traffic accident identification method provided in an embodiment of the present application;

[0043] Figure 3 It is a schematic diagram of the implementation process of a traffic accident identification method provided in an embodiment of the present application;

[0044] Figure 4 It is a schematic diagram of determining the priority of accident handling provided by an embodiment of the present application;

[0045] Figure 5 is a schematic diagram of a traffic accident identification device provided in an embodiment of the present application;

[0046] Figure 6 It is a schematic diagram of a traffic accident identification system provided in an embodiment of the present application. Embodiments of the present invention

[0047] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0048] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

[0049] Identification of abnormal vehicle conditions during driving, including traffic accident identification, can effectively improve emergency response speed, reduce traffic congestion, and improve road safety. By real-time monitoring and automatic identification of traffic accidents, emergency services can be alerted quickly, shortening response time and saving lives at critical moments; by quickly identifying and responding to traffic accidents, the time spent at the scene of the accident can be reduced, thereby reducing the resulting traffic congestion; traffic accident identification technology can monitor and analyze traffic flow, predict potential traffic accidents, and alert drivers, especially fleet drivers, through early warning systems, thereby reducing the risk of accidents. Therefore, accurate identification of traffic accidents is of great significance.

[0050] Figure 1 A schematic diagram of an implementation scenario of a traffic accident identification method provided in an embodiment of the present application. Figure 1 As shown, the implementation scenario includes a vehicle end 1 and a platform end 2, wherein the vehicle end 1 is used to collect multimodal data and perform preliminary screening. The equipment used to collect multimodal data may include a driver monitoring system (DMS in English, Driver Monitoring System in English) 10, an advanced driver assistance system 11 (ADAS in English, Advanced Driver Assistance System in English), and a motion sensor 12. Among them, the driver monitoring system 10 can collect relevant information about the driver of the vehicle through the DMS camera 101, determine the driver's action or expression of the vehicle, including at least one of abnormal driver movement and driver panic, and the driver monitoring system 10 can use a sound sensor 102 to detect the voice in the cockpit, detect abnormal screams of the driver and other information. The advanced driver assistance system 11 can collect the environment during the vehicle driving process through the ADAS camera, and detect whether there is an abnormality in front of the vehicle, including at least one of vehicle collision and pedestrian collision. The motion sensor 12 can be used to collect the vehicle's motion information and determine the vehicle's motion state, including at least one of abnormal rapid deceleration and abnormal vehicle posture. The platform end 2 can perform preliminary screening of suspected traffic accidents based on the collected multimodal data, and send the multimodal data of the suspected traffic accidents to the platform end 2. The platform end 2 obtains the identification result of the suspected traffic accident based on the driver's personalized driving habits, historical accident information of the location of the suspected traffic accident and the multimodal data.

[0051] Figure 2 A schematic diagram of the implementation flow of a traffic accident identification method provided in an embodiment of the present application is described in detail as follows:

[0052] In S201, the vehicle side acquires multimodal data, and performs preliminary screening of traffic accidents based on the multimodal data to acquire information related to suspected traffic accidents.

[0053] Among them, the multimodal data includes the driver information on the vehicle side, and the information related to the suspected traffic accident includes the location of the suspected traffic accident.

[0054] The multimodal data may include image data outside the cockpit, image data inside the cockpit, voice data inside the cockpit, and vehicle motion data. Figure 1 The device for collecting multimodal data shown includes ADAS, DMS and motion sensor. Among them, ADAS can be used to obtain image data and audio data in the cockpit. DMS can be used to obtain image data outside the cockpit. The motion state data of the vehicle is obtained through the motion sensor.

[0055] Among them, when the vehicle side determines the relevant information of the suspected traffic accident based on the multimodal data, it can make judgments on the suspected traffic accident based on the collected multimodal data.

[0056] For example, at least one of abnormal driver movement and driver panic can be detected based on the in-cockpit image data collected by ADAS.

[0057] The speed of the driver's abnormal movement can be determined based on the driver's position in the image and the frame rate of the image. If both the abnormal movement speed and the movement amplitude meet the set requirements, a suspected traffic accident based on the driver's abnormal movement can be determined. For example, the moving distance and moving speed of the driver's head can be detected. If the moving speed is greater than a predetermined speed threshold, and the moving distance is greater than a predetermined amplitude threshold, a suspected traffic accident due to the driver's abnormal movement can be determined.

[0058] The driver's expression can be obtained based on the collected image data in the cockpit. When the driver's expression is identified as a panic expression through the model, a suspected traffic accident determined based on the driver's panic is obtained.

[0059] The image data outside the cockpit collected by the DMS can be used to detect whether the vehicle has at least one of a vehicle collision and a pedestrian collision.

[0060] Based on the image data outside the cockpit collected by the DMS camera, the distance between the vehicle and other vehicles or the distance between the vehicle and pedestrians can be determined, and based on the distance, it can be determined whether the vehicle has an abnormal vehicle collision or behavior collision. Alternatively, the distance between the vehicle and other objects can also be determined based on distance detection equipment such as laser radar.

[0061] The motion data of the vehicle can be collected by the motion sensor to detect whether the vehicle is in an abnormal deceleration state or in a state of suspected traffic accident due to abnormal vehicle posture. Abnormal vehicle posture includes vehicle rollover, abnormal vehicle pitch angle, etc.

[0062] Based on motion sensors, including gyroscopes or acceleration sensors, the deceleration of the vehicle in the direction of travel can be detected. If the deceleration is less than a predetermined speed threshold, a suspected traffic accident based on abnormal deceleration can be determined. The posture of the vehicle can be detected by a gyroscope or other six-axis sensor, and based on the size of the angle between the vehicle and the standard state, it can be determined whether it is in a suspected traffic accident state of abnormal vehicle posture, including such as vehicle rollover, abnormal vehicle pitch angle, etc.

[0063] Not limited to this, you can also combine different multimodal data for comprehensive judgment, such as:

[0064] The driver's abnormal movement can be combined with expression recognition, and the image data in the cockpit collected by ADAS can be used to detect the speed and amplitude of the driver's head movement. If the movement speed and amplitude exceed the set threshold, it can be judged as abnormal movement. At the same time, combined with expression recognition technology, it can be analyzed whether the driver shows emotions such as panic. If both conditions are met at the same time, it can be determined as a suspected traffic accident.

[0065] Alternatively, the forward collision can be combined with the driver's status to detect whether there is a risk of vehicle collision or pedestrian collision through the image data outside the cockpit collected by the DMS. At the same time, combined with the driver's status (such as abnormal movement or frightened expression), the possibility of collision can be further confirmed. For example, if a pedestrian is detected in front and the driver shows panic, it can be more efficiently judged as a high-risk state.

[0066] Alternatively, the motion sensor can be combined with the driver's behavior to monitor the dynamic changes of the vehicle (such as deceleration, rollover, etc.) through the motion sensor, and combined with the driver's abnormal behavior (such as violent movement or panic expression), it can be comprehensively judged whether there is a risk of traffic accidents. For example, if the vehicle decelerates abnormally and the driver shows panic, it can be considered that there is a potential risk of an accident.

[0067] In S202, the vehicle side sends the multimodal data and the information related to the suspected traffic accident to the platform side.

[0068] Due to the limitation of computing power factors on the vehicle side, the accuracy of suspected traffic accidents determined by multimodal features is limited. After the vehicle side determines the suspected traffic accident based on multimodal data, the embodiment of the present application sends the information related to the suspected traffic accident in combination with the multimodal data corresponding to the suspected traffic accident to the platform side for further detailed analysis and processing, so as to obtain a more accurate traffic accident identification result, reduce the false alarm rate, and improve the vehicle's driving efficiency.

[0069] Among them, after determining the suspected traffic accident, the vehicle side can obtain the vehicle's location information as the location of the suspected traffic accident of the vehicle.

[0070] The vehicle's location information can be determined through satellite positioning signals, or based on lane auxiliary positioning base stations, including radio frequency positioning base stations to determine the location of suspected traffic accidents, or the location of suspected traffic accidents can be determined based on vehicle communication modules, or the location of suspected traffic accidents can be determined through other vehicles through the in-vehicle interconnection communication system.

[0071] In a possible implementation, after the embodiment of the present application sends the suspected traffic accident related information and multimodal data to the platform, the vehicle side can further collect multimodal data for a predetermined time and send the multimodal data for the predetermined time to the platform, so that the platform can obtain more accurate analysis results. For example, the vehicle side can further collect multimodal data for 1-5 minutes and send it to the platform.

[0072] In S203, the platform determines the driver's personalized driving habits based on the driver information of the vehicle, and determines the historical accident information of the location based on the location of the suspected traffic accident.

[0073] The driver information may include the driver's characteristic data in the cockpit image data collected by the ADAS. For example, the platform can determine the driver's face image based on the cockpit image data collected by the ADAS, search for the driver's identification based on the driver's face image, and determine the driver's personalized driving habits based on the driver identification found.

[0074] The driving habit data includes habit data such as habit of closing eyes, habit of sudden braking, habit of sudden acceleration, etc., which are second related features associated with driving habits.

[0075] According to the time when the suspected traffic accident was detected, combined with the collected positioning information, the historical accident information of the location of the suspected traffic accident is determined. The historical accident information matching the location of the suspected traffic accident can be determined by querying the accident database based on the location, for example, the historical accident information whose distance from the location of the suspected traffic accident is less than a predetermined distance radius.

[0076] The historical accident information searched includes the main accident factors that caused the historical accident, that is, the first relevant features related to the historical accident information, including main accident factors such as road surface factors, visibility factors, driver operation factors, etc.

[0077] Based on the determined first related features and second related features, the weights of the features in the collected multimodal data may be adjusted according to a proportional correspondence relationship, including the weight of the first related feature according to the frequency of historical traffic accidents at the location where the suspected traffic accident was detected. That is, the higher the frequency of historical traffic accidents at the location, the higher the weight of the first related feature, so that the traffic accident occurring at the location can be more reliably identified.

[0078] According to the degree of habit of the driver's personalized driving habits, the weight of the second related feature is adjusted in accordance with an inverse proportional correspondence. For example, the higher the degree of habit of the personalized driving habits, the lower the weight of the second related feature, thereby reducing the number of false alarms that are recognized as traffic accidents due to the personalized driving habits, which is conducive to improving the recognition accuracy.

[0079] In S204, the platform obtains an identification result of a suspected traffic accident based on the driver's personalized driving habits, historical accident information of the location, and the multimodal data.

[0080] In the embodiment of the present application, the multimodal data uploaded by the platform end may include multimodal data within a certain time period before the suspected traffic accident is detected, or may also include multimodal data within a predetermined time period after the suspected traffic accident is detected. The certain time period before the suspected traffic accident is detected and the predetermined time period after the suspected traffic accident is detected may be the same time period or different time periods.

[0081] In the embodiment of the present application, the platform can generate an accident description text based on the information related to the suspected traffic accident, the driver's personalized driving habits, and the historical accident information of the location where the suspected traffic accident was detected. The accident description text can be recognized by a large language model to make a more detailed analysis and judgment of the suspected traffic accident, thereby obtaining an identification result of the suspected traffic accident.

[0082] like Figure 3 The figure shows a schematic diagram of the implementation process of accident identification using a large language model provided by an embodiment of the present application. In the implementation process, the driver identification can be searched based on the image data in the cockpit, and the driver's historical driving data can be searched based on the driver identification. The driver's historical driving data can be input into a trained habit data extraction model to obtain the driver's personalized driving habits. According to the location of the suspected traffic accident detected, the traffic accident database is searched for historical traffic accidents whose distance from the location of the suspected traffic accident meets a predetermined distance radius, that is, the distance from the location of the suspected traffic accident is less than the distance radius, and the historical traffic accident information related to the historical traffic accident searched is obtained.

[0083] Based on the uploaded multimodal data, combined with the determined historical traffic accident information and personalized driving habits, textual accident description information can be obtained. The accident description information is input into the language model to obtain the identification result of the suspected traffic accident based on the analysis of the language large model.

[0084] Since this method is based on historical traffic accident information determined based on the location of a suspected traffic accident, combined with the driver's personalized driving data for analysis and judgment, it is conducive to obtaining more accurate accident identification results, reducing false alarms, and improving fleet driving efficiency.

[0085] In a possible implementation, there are a large number of suspected traffic accidents reported by vehicles. For example, the platform may receive information related to suspected traffic accidents uploaded by N vehicles. In order to give priority to responding to more urgent suspected traffic accidents, the following can be adopted: Figure 4 The method for determining the priority of suspected traffic accidents shown includes:

[0086] In S401, the platform determines the accident level according to the six-axis acceleration and the vehicle speed acquired by the sensor and the weight information corresponding to the terrain at the current position of the vehicle.

[0087] The embodiment of the present application can determine the six-axis acceleration and driving speed of the vehicle based on the motion data collected by the motion sensor on the vehicle side. Among them, the six-axis acceleration acquisition device of the vehicle includes a three-axis accelerometer and a three-axis gyroscope. The three-axis accelerometer is used to measure the linear acceleration of the vehicle in the three directions of X, Y, and Z (generally, the forward direction of the vehicle is the X direction, and the vertical upward direction is the Z direction), while the three-axis gyroscope measures the angular velocity of the vehicle around the three axes of X, Y, and Z. Including: the X-axis gyroscope is used to measure the rotation speed of the vehicle around the X-axis, that is, the yaw angular velocity of the vehicle; the Y-axis gyroscope is used to measure the rotation speed of the vehicle around the Y-axis, that is, the pitch angular velocity of the vehicle; the Z-axis gyroscope is used to measure the rotation speed of the vehicle around the Z-axis, that is, the roll angular velocity of the vehicle.

[0088] Among them, when the weight information is determined according to the terrain at the current position of the vehicle, the terrain information may include off-road terrain, non-hardened road terrain and hardened road terrain. Among them, the lower the flatness of the road surface, the weight of the acceleration in the Z direction and the rotation speed along the X axis and the Y axis can be reduced accordingly, and the weight of the vehicle speed will be increased accordingly. The weight of the linear acceleration in the X-axis direction and the rotation speed along the Y axis is adjusted according to the inclination of the road surface. For example, the greater the inclination of the road surface, the smaller the weight of the rotation speed along the Y axis.

[0089] Based on the detected six-axis acceleration and vehicle speed, combined with the weight determined according to the terrain, a priority score of the suspected traffic accident is obtained, and the accident level of each suspected traffic accident is determined according to the priority score.

[0090] In S402, the platform determines the priority of handling the vehicle accident according to the accident level.

[0091] Based on the determined accident level, suspected traffic accidents with higher levels are handled with priority, thereby enabling more reliable identification of more serious traffic accidents.

[0092] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0093] Figure 5 A schematic diagram of a traffic accident identification device provided in an embodiment of the present application, the device comprising:

[0094] The data acquisition unit 501 is used to acquire multimodal data from the vehicle side, and perform preliminary screening of traffic accidents based on the multimodal data to acquire information related to suspected traffic accidents. The multimodal data includes driver information on the vehicle side, and the information related to suspected traffic accidents includes the location of the suspected traffic accident.

[0095] The information sending unit 502 is used for sending the multimodal data and the suspected traffic accident related information to the platform side from the vehicle side.

[0096] The information determination unit 503 is used for the platform to determine the driver's personalized driving habits based on the driver information of the vehicle, and to determine the historical accident information of the location based on the location of the suspected traffic accident.

[0097] The accident identification unit 504 is used for the platform to obtain an identification result of a suspected traffic accident based on the driver's personalized driving habits, the historical accident information of the location and the multimodal data.

[0098] Figure 5 The traffic accident identification device shown in Figure 2 The traffic accident identification method shown corresponds to.

[0099] Figure 6 Schematic diagram of a traffic accident identification system provided by an embodiment of the present application. Figure 6As shown, the traffic accident identification system 6 of this embodiment includes: a processor 60, a memory 61, and a computer program 62 stored in the memory 61 and executable on the processor 60, such as a traffic accident identification program. When the processor 60 executes the computer program 62, the steps in the above-mentioned traffic accident identification method embodiments are implemented. Alternatively, when the processor 60 executes the computer program 62, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0100] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 62 in the traffic accident identification system 6.

[0101] The traffic accident identification system may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will appreciate that Figure 6 It is only an example of a traffic accident identification system 6 and does not constitute a limitation of the traffic accident identification system 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the traffic accident identification system may also include input and output devices, network access devices, buses, etc.

[0102] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0103] The memory 61 may be an internal storage unit of the traffic accident identification system 6, such as a hard disk or memory of the traffic accident identification system 6. The memory 61 may also be an external storage device of the traffic accident identification system 6, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the traffic accident identification system 6. Further, the memory 61 may also include both an internal storage unit and an external storage device of the traffic accident identification system 6. The memory 61 is used to store the computer program and other programs and data required by the traffic accident identification system. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0104] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0105] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0106] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0107] In the embodiments provided in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0108] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0109] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0110] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process in the above-mentioned embodiment method, and can also be completed by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0111] In addition, an embodiment of the present application also provides a computer program product, which, when executed on a computer, enables the computer to execute the methods in the above-mentioned implementation modes.

[0112] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A traffic accident identification method, characterized in that: The method comprises: The vehicle side acquires multimodal data, and performs preliminary screening of traffic accidents based on the multimodal data to acquire information related to suspected traffic accidents, wherein the multimodal data includes driver information of the vehicle side, and the information related to suspected traffic accidents includes the location of the suspected traffic accident; The vehicle end sends the multimodal data and the suspected traffic accident related information to the platform end; The platform determines the driver's personalized driving habits based on the driver information of the vehicle, and determines the historical accident information of the location based on the location of the suspected traffic accident; The platform obtains the identification result of the suspected traffic accident according to the driver's personalized driving habits, the historical accident information of the location and the multimodal data.

2. The method according to claim 1, characterized in that The platform obtains the identification result of the suspected traffic accident according to the personalized driving habits of the driver, the historical accident information of the location and the multimodal data, including: The platform generates an accident description text based on the information related to the suspected traffic accident, the driver's personalized driving habits, and the historical accident information of the location; The platform inputs the accident description text into a large language model to obtain a recognition result of the suspected traffic accident.

3. The method according to claim 1, characterized in that The vehicle side sends the multimodal data and the suspected traffic accident related information to the platform side, and the method further includes: After the vehicle end obtains the information related to the suspected traffic accident, the vehicle end continues to obtain multimodal data for a predetermined time period; The vehicle side sends the multimodal data used to determine the information related to the suspected traffic accident, the multimodal data that continues to be acquired, and the traffic accident information to the platform side.

4. The method according to claim 1, characterized in that: After the platform obtains the identification result of the suspected traffic accident according to the personalized driving habits of the driver, the historical accident information of the location and the multimodal data, the method further includes: The platform determines the accident level based on the vehicle's six-axis acceleration and vehicle speed acquired by the sensor and the weight information corresponding to the terrain at the vehicle's current location; The platform determines the priority of handling the vehicle accident according to the accident level.

5. The method according to claim 1, characterized in that The multimodal data includes vehicle motion data, in-cockpit image data, and out-cockpit image data; Perform preliminary screening of traffic accidents based on the multimodal data to obtain relevant information of suspected traffic accidents, including: The vehicle end determines at least one of a sudden deceleration of the vehicle, a collision of the vehicle, and a rollover of the vehicle according to the vehicle motion data; The vehicle end determines at least one of an abnormal driver movement event and a driver panic event based on the in-cockpit image data.

6. The method according to any one of claims 1 to 5, characterized in that: The platform obtains the identification result of the suspected traffic accident based on the driver's personalized driving habits, the historical accident information of the location and the multimodal data, including: Determining a first relevant feature associated with historical accident information, and determining a second relevant feature associated with personalized driving habits; The weight of the first relevant feature is determined according to the frequency of historical accident information at the location, and the weight of the second relevant feature is determined according to the personalized driving habit of the driver.

7. The method according to claim 6, characterized in that Determining the weight of the first relevant feature according to the frequency of the historical accident information of the location includes: According to the frequency of historical traffic accidents at the location, the weight of the first relevant feature is adjusted in a proportional relationship; determining a weight of the second relevant feature according to the personalized driving habits of the driver; According to the degree of the driver's personalized driving habit, the weight of the second related feature is adjusted in an inverse proportional correspondence.

8. A traffic accident identification device, characterized in that: The device comprises: A data acquisition unit, configured to acquire multimodal data from a vehicle, and perform preliminary screening of traffic accidents based on the multimodal data to acquire information related to suspected traffic accidents, wherein the multimodal data includes information about the driver of the vehicle, and the information related to suspected traffic accidents includes the location of the suspected traffic accident; An information sending unit, configured to send the multimodal data and the suspected traffic accident related information to the platform end from the vehicle end; An information determination unit, configured to determine the driver's personalized driving habits according to the driver information of the vehicle end by the platform end, and determine the historical accident information of the location according to the location of the suspected traffic accident; The accident identification unit is used for obtaining the identification result of the suspected traffic accident by the platform according to the personalized driving habits of the driver, the historical accident information of the location and the multimodal data.

9. A traffic accident identification system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the traffic accident identification system implements the method according to any one of claims 1 to 7.

10. A computer program product comprising computer program instructions, characterized in that When the computer program is executed, the method according to any one of claims 1 to 7 is performed.

Images