Vehicle collision identification method and device, storage medium and electronic equipment

Abstract

The invention relates to a vehicle collision identification method and device, a storage medium and electronic equipment, and relates to the technical field of collision identification, and the method comprises the steps: determining first collision information according to GPS data, determining a recorder video as second collision information, and according to the first collision information and the second collision information, carrying out vehicle collision identification; determining whether the target vehicle has a collision risk; if yes, according to the in-vehicle video, determining an actual dangerous driving operation of the target driver within a preset duration before the current time, and determining an actual driving condition before the current time of the target vehicle; and according to each actual dangerous driving operation, the actual driving condition, at least one target dangerous driving operation easy to occur by the target driver and the corresponding target driving condition, verifying whether the target vehicle has a collision risk, and if yes, sending a safety early warning to a terminal of a collision verification person. The method and the device have the effect of improving the timeliness of rescuing the personnel with the collision accident.

Description

Technical Field

[0001] The present application relates to the technical field of collision recognition, and in particular to a vehicle collision recognition method, device, storage medium and electronic device. Background Art

[0002] In recent years, with the continuous development of China's economy and the improvement of people's living standards, cars have gradually become a necessity for more and more families, and domestic consumers' enthusiasm for car consumption has only increased. As of the end of September 2023, the national motor vehicle ownership has reached 430 million, of which cars still account for a dominant proportion, reaching 330 million. This figure has increased significantly compared with previous years, reflecting the huge scale and strong demand of China's automobile market. Against this background, road traffic congestion in various cities has also intensified, resulting in an increasing number of vehicle collision accidents. The negative impacts of vehicle collision accidents are also multifaceted, not only involving economic losses, but also may have a profound impact on personal health, social resources and the environment. Therefore, there is an urgent need for an effective monitoring mechanism or effective identification mechanism for vehicle collision accidents.

[0003] At present, the common method for identifying vehicle collision accidents is: after a vehicle collision accident occurs, it is mainly based on the accident report filed by the vehicle owner and related personnel, or the corresponding safety customer service personnel of the car factory will take the initiative to call to inquire about the relevant circumstances of the accident and the safety status of the vehicle owner when monitoring the abnormality of the vehicle. After confirming the occurrence of a collision accident, a targeted response will be made. Under this method, the identification and confirmation of vehicle collision accidents involves telephone communication, which takes a long time. Once the collision accident is more serious, this method is less timely for the rescue of the people involved in the collision accident. Summary of the invention

[0004] In order to improve the timeliness of rescue of people involved in a collision accident, the present application provides a vehicle collision identification method, device, storage medium and electronic device.

[0005] In a first aspect of the present application, a vehicle collision recognition method is provided, which specifically includes: When the target vehicle vibrates at the current time, the recorder video and GPS data of the target vehicle before and after the current time are obtained, and the in-vehicle video of the target vehicle within a preset time period before the current time is obtained, and the GPS data is used to characterize the location information and speed information of the target vehicle; Determine first collision information according to the GPS data, determine the recorder video as second collision information, and determine whether the target vehicle has a collision risk according to the first collision information and the second collision information; If yes, then determining at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determining the actual driving condition of the target vehicle before the current time; Based on the actual dangerous driving operations, the actual driving conditions, at least one target dangerous driving operation that is prone to occur by the target driver, and the corresponding target driving conditions, it is verified whether the target vehicle has a risk of collision. If so, a safety warning is sent to the terminal of the collision verification personnel. The target driving conditions are driving conditions that are likely to cause collision accidents when the corresponding target dangerous driving operations are present.

[0006] By adopting the above technical solution, when the target vehicle vibrates, the recorder video, GPS data and in-car video are obtained. Then, the speed change of the target vehicle during driving is analyzed through GPS data to determine the first collision information. At the same time, the recorder video is determined as the second collision information. The first collision information and the second collision information are combined to determine whether the target vehicle has a risk of collision. If there is a risk of collision, based on the target dangerous driving operation and the corresponding target driving condition, combined with the actual dangerous driving operation and actual driving condition of the target driver, the possibility of the target vehicle colliding with the current target vehicle is analyzed, and then it is verified whether the target vehicle does have a risk of collision, so as to more accurately evaluate the collision situation of the target vehicle. Finally, after verifying that there is a risk of collision, the target vehicle is likely to have a collision accident, and a safety warning is directly sent to the collision verification personnel. Once the target vehicle collides, the personnel can go to the scene in time to provide rescue, thereby improving the timeliness of rescue for people in collision accidents.

[0007] Optionally, verifying whether the target vehicle has a risk of collision based on the actual dangerous driving operations, the actual driving conditions, at least one target dangerous driving operation that the target driver is prone to, and the corresponding target driving conditions, specifically includes: Acquire historical dangerous driving operations of the driver in the vehicle when the vehicle collided historically, count the first occurrence times of each of the historical dangerous driving operations, and select a first number of historical dangerous driving operations from each of the historical dangerous driving operations in descending order of the first occurrence times to determine as target dangerous driving operations; Acquire historical driving conditions of the vehicle when a collision accident occurred under a single target dangerous driving operation by a historical driver, count the second number of occurrences of each of the historical driving conditions, and select a second number of historical driving conditions from each of the historical driving conditions in descending order of the second number of occurrences to determine as target driving conditions for the corresponding target dangerous driving operation; Calculating a first weight for each of the target dangerous driving operations, and calculating a second weight for each of the target driving conditions corresponding to each of the target dangerous driving operations, wherein the first weight is a ratio of a first occurrence number of each target dangerous driving operation to a sum of first occurrence numbers of all target dangerous driving operations, and the second weight is a ratio of a second occurrence number of a single target driving condition corresponding to the target dangerous driving operation to a sum of second occurrence numbers of all corresponding target driving conditions; Based on each of the actual dangerous driving operations, the actual driving conditions, the first weights and the corresponding second weights, it is verified whether the target vehicle has a risk of collision.

[0008] By adopting the above technical solution, the larger the first number of occurrences, the more likely the corresponding historical dangerous driving operation is to cause a collision accident, and then the target dangerous driving operation is determined; the larger the second number of occurrences, the more likely the vehicle is to cause a collision accident in the corresponding historical driving condition when the target dangerous driving operation occurs during driving, and then the corresponding target driving condition is determined. Finally, combined with the first weight of the target dangerous driving operation and the second weight of each corresponding target driving condition, the probability of the target vehicle colliding at the current time is analyzed and determined, and then the verification of whether the target vehicle has a risk of collision is completed, and the accurate assessment of the collision situation of the target vehicle is achieved.

[0009] Optionally, verifying whether the target vehicle has a risk of collision based on each of the actual dangerous driving operations, the actual driving conditions, the first weights and the corresponding second weights specifically includes: Determine the target dangerous driving operation that exists in the actual driving condition in the corresponding target driving conditions as the key driving operation, and when the actual dangerous driving operation is the key driving operation, determine the corresponding actual dangerous driving operation as the reference driving operation; Calculating a first product of a first weight of each reference driving operation and a second weight of a corresponding actual driving condition, and summing the first products to obtain a sum of the first products; If the sum of the first products is greater than a preset product sum threshold, it is verified that there is a risk of collision with the target vehicle.

[0010] By adopting the above technical solution, the larger the first product is, the greater the possibility of a collision accident when the target driver makes the corresponding reference driving operation, and the larger the sum of the first products is, the greater the possibility of a collision accident currently occurring for the target driver. Finally, if the sum of the first products is greater than the product sum threshold, it means that the possibility of a collision accident currently occurring is high, and then it is verified that the target vehicle does have a collision risk, thereby more accurately assessing the collision situation of the target vehicle.

[0011] Optionally, the method further includes: If the target vehicle does not have a collision risk, determining the current driving condition of the target vehicle, and determining the target dangerous driving operation that exists in the current driving condition in each corresponding target driving condition as an important driving operation; Calculating a second product of the first weight of each of the important driving operations and the second weight of the corresponding current driving condition, and summing the second products to obtain a sum of the second products; If the sum of the second products is greater than a preset product sum threshold, when the target driver performs at least one important driving operation, the corresponding important driving operation is determined as a risky driving operation, and the important driving operations other than the risky driving operation are determined as remaining driving operations; A maximum second product is selected from the second products corresponding to the remaining driving operations, and the remaining driving operation corresponding to the maximum second product is determined as a vigilant driving operation, and a reminder message is sent to the vehicle-mounted terminal of the target vehicle for the vigilant driving operation.

[0012] By adopting the above technical solution, the larger the second product is, the more likely it is to cause a collision accident when the target driver makes the corresponding important driving operation under the current driving condition, and the larger the sum of the second products is, the more likely it is that the target vehicle will have a collision accident under the current driving condition. If the sum of the second products is greater than the product sum threshold, it means that the target vehicle is more likely to have a collision accident under the current driving condition, and the target driver needs to be vigilant about his driving operation. Furthermore, the remaining driving operations corresponding to the maximum second product have a greater impact on the occurrence of an acceleration collision accident, so they are determined as vigilant driving operations, and a reminder message is sent to the on-board terminal of the target vehicle for this vigilant driving operation, so as to promptly remind the target driver to avoid this vigilant driving operation, thereby better avoiding the occurrence of a collision accident.

[0013] Optionally, the method further includes: If the sum of the second products is not greater than a preset product-sum threshold, calculating a third product of the first weight of each of the risky driving operations and the second weight of each corresponding target driving condition; Summing the third products corresponding to the same target driving condition to obtain a corresponding sum of the third products, and selecting a maximum sum of the third products from the sums of the third products; When the sum of the maximum third products is greater than a preset product sum threshold, the target driving condition corresponding to the maximum sum of the third products is determined as a vigilant driving condition; if the vigilant driving condition is different from the current driving condition, a reminder message is sent to the on-board terminal of the target vehicle for the vigilant driving condition.

[0014] By adopting the above technical solution, if the sum of the second products is not greater than the product sum threshold, it means that the target vehicle is less likely to have a collision accident under the current driving condition. Furthermore, the larger the sum of the third products is, the greater the possibility of a collision accident under the corresponding target driving condition after the current time. If the sum of the maximum third products is greater than the product sum threshold, it means that the possibility of a collision accident under the target driving condition corresponding to the maximum sum of the third products is greater, and the target driving condition is determined as a vigilant driving condition. Furthermore, if this vigilant driving condition is different from the current driving condition, then a reminder message is sent to the on-board terminal of the target vehicle for this vigilant driving condition, thereby reminding the target driver to avoid switching to this vigilant driving condition at present to avoid the occurrence of a collision accident.

[0015] Optionally, the method further includes: Acquire historical dangerous driving operations of the driver in the vehicle when the vehicle collided historically, count the first occurrence times of each of the historical dangerous driving operations, and select a first number of historical dangerous driving operations from each of the historical dangerous driving operations in descending order of the first occurrence times to determine as target dangerous driving operations; Obtaining historical regions where collision positions on the vehicle are located when collision accidents occur under a single target dangerous driving operation by historical drivers, counting the third number of occurrences of each of the historical regions, and selecting a third number of historical regions from each of the historical regions in descending order of the third number of occurrences to determine a target region corresponding to the target dangerous driving operation; Calculating a first weight for each of the target dangerous driving operations, and calculating a third weight for each of the target areas corresponding to each of the target dangerous driving operations, wherein the first weight is a ratio of a first occurrence number of each target dangerous driving operation to a sum of first occurrence numbers of all target dangerous driving operations, and the third weight is a ratio of a third occurrence number of a single target area corresponding to the target dangerous driving operation to a sum of third occurrence numbers of all corresponding target areas; During the driving process of the target vehicle, if the target driver performs a target dangerous driving operation, the corresponding target dangerous driving operation is determined as an operation to be recorded, and a mapping relationship between each of the operations to be recorded and a corresponding occurrence time node in a real-time video is established, where the real-time video is a driving video obtained in real time from a driving recorder in the target vehicle; When the target vehicle switches from a driving state to a normal parking state and there are unknown collision marks, the cause of the unknown collision marks is checked based on the mapping relationships, the first weights and the corresponding third weights.

[0016] By adopting the above technical solution, the larger the first number of occurrences, the more likely the corresponding historical dangerous driving operation is to cause a collision accident, thereby determining the target dangerous driving operation; the larger the third number of occurrences, the more likely the location of the collision is to be in the corresponding historical area under the premise of making a single target dangerous driving operation, thereby determining the target area, and finally combining the mapping relationship, the first weight of the target dangerous driving operation and the corresponding third weight of each target area, to conduct targeted investigation of the corresponding time nodes in the real-time video, thereby more efficiently and accurately investigating the cause of the occurrence.

[0017] Optionally, the step of troubleshooting the cause of the unknown collision trace based on the mapping relationships, the first weights and the corresponding third weights specifically includes: Determine the target area where the unknown collision trace is located as a key area, determine the target dangerous driving operation in the key area in each corresponding target area as an induced collision operation, determine the operation to be recorded in each of the induced collision operations as a final operation, and determine the corresponding final time node according to the mapping relationship of the final operation; Calculating a fourth product of the first weight of each of the final operations and the third weight of the corresponding key area; According to each of the fourth weights, the playback order of the corresponding final time nodes in the real-time video is determined, and based on each of the playback orders, the cause of the occurrence of the unknown collision traces is investigated. The larger the fourth product is, the earlier the corresponding playback order is.

[0018] By adopting the above technical solution, the larger the fourth product is, the more likely it is that the unknown collision mark will appear when the target driver makes the corresponding final operation and causes a collision. Then, the earlier the playback order of the corresponding final time node in the real-time video is, the easier it is to find the cause of the unknown collision mark. Finally, according to each playback order, personnel can screen the real-time video in a targeted manner and quickly find out the cause.

[0019] In a second aspect of the present application, a vehicle collision recognition device is provided, specifically comprising: A data acquisition module, used to acquire the recorder video and GPS data of the target vehicle before and after the current time when the target vehicle vibrates at the current time, and acquire the in-vehicle video of the target vehicle within a preset time period before the current time, wherein the GPS data is used to characterize the position information and speed information of the target vehicle; a risk determination module, configured to determine first collision information according to the GPS data, determine the recorder video as second collision information, and determine whether the target vehicle has a collision risk according to the first collision information and the second collision information; a working condition determination module, for determining, if yes, at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determining the actual driving working condition of the target vehicle before the current time; The risk warning module is used to verify whether the target vehicle has a collision risk based on the actual dangerous driving operations, the actual driving conditions, at least one target dangerous driving operation that the target driver is prone to, and the corresponding target driving conditions. If so, a safety warning is sent to the terminal of the collision verification personnel. The target driving conditions are driving conditions that are likely to cause collision accidents when the corresponding target dangerous driving operations are present.

[0020] By adopting the above technical solution, the data acquisition module obtains the recorder video, GPS data and in-vehicle video. The risk determination module determines whether the target vehicle has a risk of collision based on the first collision information and the second collision information. Then, when the operating condition determination module determines that there is a risk of collision, it determines the actual dangerous driving operation and the actual driving condition. Finally, the risk warning module verifies whether the target vehicle has a risk of collision. If so, a safety warning is sent to the terminal of the collision verification personnel.

[0021] In a third aspect of the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is loaded and executed by a processor, the method steps described in any one of the first aspects are performed.

[0022] In a fourth aspect of the present application, an electronic device is provided, specifically comprising: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the processor is used to load and execute the computer program stored in the memory so that the electronic device performs the method as described in any one of the first aspects.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: analyzing the speed change of the target vehicle during driving through GPS data, and then determining the first collision information, and at the same time determining the recorder video as the second collision information, combining the first collision information and the second collision information to determine whether the target vehicle has a risk of collision, and if there is a risk of collision, based on the target dangerous driving operation and the corresponding target driving condition, combined with the target driver's actual dangerous driving operation and actual driving condition, analyzing the possibility of the target vehicle colliding with the current target vehicle, and then verifying whether the target vehicle does have a risk of collision, so as to more accurately assess the collision situation of the target vehicle. Finally, after verifying that there is a risk of collision, the target vehicle is likely to have a collision accident, then a safety warning is directly sent to the collision verification personnel, and once the target vehicle collides, the personnel can go to the scene in time to provide rescue, thereby improving the timeliness of rescue for people involved in collision accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of a vehicle collision identification method provided in an embodiment of the present application; Figure 2 It is a flow chart of another vehicle collision identification method provided in an embodiment of the present application; Figure 3 is a structural schematic diagram of a vehicle collision identification device provided in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of another vehicle collision recognition device provided in an embodiment of the present application.

[0025] Explanation of the accompanying drawings: 11. Data acquisition module; 12. Risk determination module; 13. Working condition determination module; 14. Risk warning module; 15. Operation reminder module; 16. Working condition reminder module; 17. Cause investigation module. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0027] In the description of the embodiments of the present application, words such as "illustrative", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "illustrative", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "illustrative", "for example" or "for example" is intended to present related concepts in a concrete way.

[0028] In the description of the embodiments of the present application, the term "and / or" is only a kind of association relationship describing the associated objects, indicating that there may be three kinds of relationships, for example, A and / or B, which can represent: A exists alone, B exists alone, and A and B exist at the same time. In addition, unless otherwise specified, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" can explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0029] See also Figure 1 The present application embodiment discloses a flow chart of a vehicle collision identification method, which can be implemented by a computer program or run on a vehicle collision identification device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application, specifically including: S101: When the target vehicle vibrates at the current time, the recorder video and GPS data of the target vehicle before and after the current time are obtained, and the in-vehicle video of the target vehicle within a preset time period before the current time is obtained.

[0030] Specifically, in an embodiment of the present application, the target vehicle is a moving vehicle for real-time collision monitoring, and a driving recorder is installed on the target vehicle. At the current time, if the vibration data of the target vehicle is obtained through the driving recorder, it means that the current target vehicle is vibrating; if the vibration data of the target vehicle is not obtained through the driving recorder, it means that the current target vehicle is not vibrating. When the target vehicle vibrates at the current time, it may be caused by a collision of the target vehicle. Then, the driving recorder video and GPS data before and after the current time are obtained through the driving recorder. Specifically, it can be the video and GPS data recorded by the driving recorder within 3 seconds before the current time and within 3 seconds after the current time. In other embodiments, it can also be the video and GPS data recorded by the driving recorder within 4 seconds before the current time and within 3 seconds after the current time. Among them, the recorder video can be a real-time image video of the front of the target vehicle recorded by the driving recorder. In other embodiments, the recorder video can also be a real-time image video of the front, back, left and right of the target vehicle recorded by the driving recorder. GPS data includes but is not limited to the latitude and longitude of the target vehicle before and after the current time and the driving speed, thereby characterizing the position information and speed information of the target vehicle.

[0031] Furthermore, the in-car video within a preset time period before the current time is obtained through the in-car camera, specifically: after the consent of the driver in the target vehicle, the camera opening permission is obtained, and the video information of the main driver in the car is obtained in real time through the camera, and the in-car video within the preset time period before the current time is retrieved from the video information. In addition, the execution subject of a vehicle collision identification method disclosed in an embodiment of the present application can be a server, and the server is wirelessly connected to the camera, the driving recorder and the vehicle-mounted terminal of the target vehicle. The server can be an independent physical server or a server cluster composed of multiple physical servers. In other embodiments, the execution subject of a vehicle collision identification method can also be the vehicle-mounted terminal of the target vehicle itself.

[0032] S102: Determine first collision information according to GPS data, determine the recorder video as second collision information, and determine whether there is a risk of collision with the target vehicle according to the first collision information and the second collision information.

[0033] Specifically, after the GPS data is determined, the GPS data is analyzed by the preset three-stage evaluation algorithm of GPS to determine the speed change of the target vehicle before and after the current time, and then determine the first collision information. In other embodiments, the speed change curve can also be drawn by the preset MATLAB tool according to the various speeds of the target vehicle before and after the current time, and then determine the first collision information. Exemplarily, the first collision information can be: three speed change stage information of uniform speed driving-speed drop-speed static. Further, the recorder video is determined as the second collision information. Finally, based on the preset YOLO algorithm, the first collision information and the second collision information are analyzed. The specific process is: the pedestrian, vehicle or obstacle in the second collision information (recorder video) is quickly identified by the YOLO algorithm. If the target is identified, and the first collision information is uniform speed driving-speed drop-speed static, then it means that the target vehicle may have collided and become static, and then determine that the target vehicle has a risk of collision.

[0034] S103: If yes, then determine at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determine the actual driving condition of the target vehicle before the current time.

[0035] Specifically, if it is determined that the target vehicle has a risk of collision, then the driving operation or driving action of the target driver in the car is identified based on the in-car video through a preset operation recognition model. The operation recognition model can be a trained convolutional neural network model or a trained recurrent neural network model. The training process is briefly described as follows: the driving video marked with the correct driving operation is input into the model as a training sample for training, and the model parameters are adjusted and optimized by the reverse gradient algorithm until the model converges, and finally the operation recognition model is obtained. This is a prior art and will not be repeated here. In other embodiments, the driving operations of the target driver when operating the steering wheel, stepping on the accelerator and braking during driving can also be obtained through steering sensors, accelerator pedal position sensors, brake pedal position sensors, etc.

[0036] Further, the identified driving operation is then compared with a plurality of preset dangerous driving operations, and the similarity is specifically calculated. If the similarity exceeds the preset similarity threshold, the actual dangerous driving operation is determined. Among them, the dangerous driving operation refers to some driving operations that have safety hazards when the driver is driving the vehicle. Exemplarily, the dangerous driving operation includes but is not limited to sudden steering, slamming on the accelerator, slamming on the brakes, and not turning on the lights correctly. Further, the driving trajectory of the target vehicle before the current time is determined by analyzing the acquired GPS data, and then the actual driving condition of the target vehicle is determined. In other embodiments, the actual driving condition of the target vehicle before the current time can also be obtained in real time through the navigation tool in the car. It should be noted that the driving condition of the vehicle includes but is not limited to starting condition, acceleration condition, turning condition, uphill and downhill condition, and reversing condition, etc. It should be noted that a feasible way to calculate the similarity is to map the identified driving operation and the dangerous driving operation to a high-dimensional vector space respectively, obtain the corresponding vector, and then calculate the cosine similarity between the two vectors, and then determine the similarity.

[0037] S104: Verify whether the target vehicle has a risk of collision based on actual dangerous driving operations, actual driving conditions, at least one target dangerous driving operation that the target driver is prone to, and the corresponding target driving conditions. If so, send a safety warning to the terminal of the collision verification personnel.

[0038] Specifically, historical news information of vehicle collision accidents is obtained through search engines or crawlers. The historical news information includes but is not limited to the driving operation of the driver at the time of the collision accident and the driving condition information, etc. Based on the historical news information, the historical dangerous driving operations of the driver in the vehicle when the historical vehicle collided are determined, and the first occurrence number of each historical dangerous driving operation is counted. The larger the first occurrence number, the more likely the corresponding historical dangerous driving operation is to cause a collision accident. Then, according to the order of the first occurrence number from large to small, the first number of historical dangerous driving operations are selected from each historical dangerous driving operation to be determined as the target dangerous driving operation, that is, the dangerous driving operation that is likely to cause a collision accident. Based on the above historical news information, the historical driving condition of the vehicle when the historical driver has a collision accident under a single target dangerous driving operation is determined, and the second occurrence number of each historical driving condition is counted. The larger the second occurrence number, the more likely the vehicle is in the corresponding historical driving condition when the target dangerous driving operation occurs during driving. According to the order of the second occurrence number from large to small, the second number of historical driving conditions are selected from each historical driving condition to be the target driving condition of the corresponding target dangerous driving operation, that is, the driving condition that is likely to cause a collision accident.

[0039] Further, the first weight of each target dangerous driving operation and the second weight of each target driving condition corresponding to each target dangerous driving operation are calculated, the first weight is the ratio of the first occurrence number of each target dangerous driving operation to the sum of the first occurrence number of all target dangerous driving operations, and the second weight is the ratio of the second occurrence number of a single target driving condition corresponding to the target dangerous driving operation to the sum of the second occurrence number of all corresponding target driving conditions. After the first weight and the corresponding second weights are determined, the existence of the current collision risk of the target vehicle is further verified in combination with the actual dangerous driving operation and the actual driving condition of the target driver within a preset time before the current time, so as to more accurately evaluate the collision situation of the target vehicle. A feasible verification method is: determine the target dangerous driving operation with the actual driving condition in the corresponding target driving conditions as the key driving operation, and when the actual dangerous driving operation is the key driving operation, determine the corresponding actual dangerous driving operation as the reference driving operation. Then, calculate the first product of the first weight of each reference driving operation and the second weight of the corresponding actual driving condition. The larger the first product, the greater the possibility of a collision accident when the target driver makes the corresponding reference driving operation. The sum of the first products is obtained by summing up the first products. The larger the sum of the first products is, the greater the possibility that the target driver is currently in a collision accident. Finally, the sum of the first products is compared with a preset product sum threshold. If the sum of the first products is greater than the product sum threshold, it means that the possibility of a collision accident is high, and it is verified that the target vehicle does have a collision risk.

[0040] Furthermore, when it is verified that the target vehicle has a risk of collision, a safety warning is sent to the terminal of the collision verification personnel. The terminal can be a personal computer or a smart phone, and the collision verification personnel can be a car repair personnel near the target vehicle, so that the collision verification personnel can go to the location of the target vehicle to investigate the accident. Once a collision accident occurs, rescue can be provided in time, and it can also help the car repair personnel to actively acquire customers.

[0041] In other embodiments, if the target vehicle does not currently have a risk of collision, then the driving condition of the target vehicle at the current time, that is, the current driving condition, is obtained, and the target dangerous driving operation in the corresponding target driving conditions that exists in the current driving condition is determined as an important driving operation, and the second product of the first weight of each important driving operation and the second weight of the corresponding current driving condition is calculated. The larger the second product is, the easier it is to cause a collision accident when the target driver performs the corresponding important driving operation under the current driving condition. Further, each second product is summed to obtain the corresponding sum of the second products. The larger the sum of the second products is, the easier it is for the target vehicle to have a collision accident under the current driving condition. If the sum of the second products is greater than the product sum threshold, it means that the target vehicle is more likely to have a collision accident under the current driving condition, and the target driver needs to be vigilant about his driving operation. Then, when the target driver currently has at least one important driving operation, in order to avoid subsequent collision accidents, it is necessary to pay attention to the driving operation of the target driver after the current time. Next, the corresponding important driving operations are determined as risky driving operations, and the important driving operations other than the risky driving operations are determined as remaining driving operations. The maximum second product is selected from the second products corresponding to the remaining driving operations. The remaining driving operations corresponding to the maximum second product have a greater impact on the occurrence of acceleration collision accidents, so they are determined as vigilant driving operations, and reminder information is sent to the on-board terminal of the target vehicle for this vigilant driving operation, so as to timely remind the target driver to avoid this vigilant driving operation, thereby better avoiding the occurrence of collision accidents.

[0042] In another embodiment, if the sum of the second products is not greater than the product-sum threshold, it indicates that the target vehicle has a low probability of a collision accident under the current driving condition, and then the third product of the first weight of each risky driving operation and the second weight of each corresponding target driving condition is calculated. Then, the third products corresponding to the same target driving condition are summed to obtain the corresponding sum of the third products. The larger the sum of the third products, the greater the probability of a collision accident of the target vehicle under the corresponding target driving condition after the current time. Then, the maximum sum of the third products is selected from the sums of the third products. If the maximum sum of the third products is greater than the product-sum threshold, it indicates that the probability of a collision accident under the target driving condition corresponding to the maximum sum of the third products is high, and then the target driving condition is determined as a vigilant driving condition. Further, if this vigilant driving condition is different from the current driving condition, then for this vigilant driving condition, a reminder message is sent to the vehicle-mounted terminal of the target vehicle, thereby reminding the target driver to avoid switching to this vigilant driving condition at present.

[0043] See also Figure 2 The present application embodiment discloses a flow chart of another vehicle collision recognition method, which can be implemented by a computer program or run on a vehicle collision recognition device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application, specifically including: S201: When the target vehicle vibrates at the current time, the recorder video and GPS data of the target vehicle before and after the current time are obtained, and the in-vehicle video of the target vehicle within a preset time period before the current time is obtained.

[0044] S202: Determine first collision information according to GPS data, determine the recorder video as second collision information, and determine whether there is a risk of collision with the target vehicle according to the first collision information and the second collision information.

[0045] S203: If yes, then determine at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determine the actual driving condition of the target vehicle before the current time.

[0046] S204: Verify whether the target vehicle has a risk of collision based on actual dangerous driving operations, actual driving conditions, at least one target dangerous driving operation that the target driver is prone to, and the corresponding target driving conditions. If so, send a safety warning to the terminal of the collision verification personnel.

[0047] For details, please refer to steps S101-S104, which will not be described in detail here.

[0048] S205: Obtain historical dangerous driving operations of the driver in the vehicle when the vehicle collided historically, count the first occurrence times of each historical dangerous driving operation, and select a first number of historical dangerous driving operations from each historical dangerous driving operation in descending order of the first occurrence times to determine as target dangerous driving operations.

[0049] For details, please refer to step S104, which will not be described in detail here.

[0050] S206: Obtain historical areas where collision positions on the vehicle are located when historical drivers have collision accidents under a single target dangerous driving operation, count the third occurrence times of each historical area, and select a third number of historical areas from each historical area in descending order of the third occurrence times to determine the target area of ​​the corresponding target dangerous driving operation.

[0051] S207: Calculate a first weight for each target dangerous driving operation, and calculate a third weight for each target area corresponding to each target dangerous driving operation.

[0052] Specifically, after the target dangerous driving operation is determined, based on the above historical news information, the historical area where the collision position on the vehicle is located when the historical driver makes a single target dangerous driving operation and a collision accident occurs is determined, wherein the historical news information also includes the collision position where the vehicle collides. The third number of occurrences of the historical area is counted, and a third number of historical areas are selected from each historical area in order of the third number of occurrences from large to small to determine the target area corresponding to the target dangerous driving operation, that is, the area on the vehicle that is prone to collision. Exemplarily, the target dangerous driving operation is changing lanes without turning on the turn signal, then the corresponding target area is usually the side rear area of ​​the vehicle. Furthermore, the first weight is the ratio of the first number of occurrences of each target dangerous driving operation to the sum of the first number of occurrences of all target dangerous driving operations, and the third weight is the ratio of the third number of occurrences of a single target area corresponding to the target dangerous driving operation to the sum of the third number of occurrences of all corresponding target areas.

[0053] S208: During the driving of the target vehicle, if the target driver performs a target dangerous driving operation, the corresponding target dangerous driving operation is determined as an operation to be recorded, and a mapping relationship between each operation to be recorded and a corresponding occurrence time node in the real-time video is established.

[0054] S209: When the target vehicle switches from the driving state to the normal parking state and there are unknown collision marks, the cause of the unknown collision marks is investigated based on the mapping relationships, the first weights and the corresponding third weights.

[0055] Specifically, when the target vehicle is in the process of driving, the driving operation of the target driver is obtained in real time through the in-vehicle camera. If the target driver performs a target dangerous driving operation, the corresponding target dangerous driving operation is determined as the operation to be recorded, and the occurrence time node of the operation to be recorded is determined through the monitoring video of the camera. Finally, a mapping relationship between each operation to be recorded and the corresponding occurrence time node in the real-time video is established, wherein the real-time video is the driving video obtained in real time from the driving recorder in the target vehicle. Further, when the target vehicle switches from the driving state to the normal parking state, it means that the target driver has finished driving. If the relevant information of the unknown collision mark sent by the terminal of the target driver is received, it means that after the driving is finished, there are collision marks on the target vehicle caused by the collision accident that was not detected during the driving process, wherein the unknown collision mark is a collision mark whose time, place and cause are uncertain during the driving of the target vehicle. For example, the unknown collision mark can be a brand new scratch. Furthermore, it is necessary to investigate the cause of the unknown collision mark. A feasible investigation method is: determine the target area where the unknown collision mark is located as the key area, determine the target dangerous driving operation in this key area in each corresponding target area as the collision-inducing operation, and then determine the operation to be recorded in each collision-inducing operation as the final operation, and then determine the corresponding final time node according to the mapping relationship of this final operation.

[0056] Further, the fourth product of the first weight of each final operation and the third weight of the corresponding key area is calculated. Finally, according to the fourth weight, the playback order of the corresponding final time node in the real-time video is determined. The larger the fourth product is, the easier it is for the unknown collision mark to appear when the target driver makes the corresponding final operation and collides, and the earlier the playback order of the corresponding final time node is. Exemplarily, the fourth weight corresponding to the final operation A is the largest, and the final time node corresponding to the final operation A is a, then the video is played from the final time node a in the real-time video first, so that personnel can quickly find the cause of the appearance of the unknown collision mark from the real-time video. Further, each final time node in the real-time video is sequentially marked according to the playback order to obtain the marked video, and finally the marked video is sent to the terminal of the target driver so that it can be marked in order, screen the real-time video in a targeted manner, and quickly find out the cause.

[0057] In other embodiments, the weight product of the first weight of at least one actual dangerous driving operation existing in each target dangerous driving operation and the third weight of each corresponding target area is calculated, and each weight product is summed to obtain the summed result of the corresponding actual dangerous driving operation, and then each summed result is summed to obtain the final summed result. If the final summed result is greater than the preset summed threshold, it means that the target vehicle is more likely to collide at the current time, then the target vehicle is verified to have a collision risk, so as to more accurately assess the collision of the target vehicle. At the same time, the weight products of the same target area are selected for summing to obtain the corresponding weight product sum, and the maximum weight product sum is selected from the weight product sums, and the target area corresponding to the maximum weight product sum is determined as the screening area, that is, the most likely collision area when a collision occurs. Finally, when sending a safety warning to the collision verification personnel, this screening area is sent together. Once a collision accident does occur, targeted rescue tools can be prepared in advance according to this screening area, thereby improving the efficiency of on-site rescue.

[0058] The implementation principle of a vehicle collision identification method in an embodiment of the present application is as follows: analyze the speed change of the target vehicle during driving through GPS data, and then determine the first collision information, and at the same time determine the recorder video as the second collision information, and combine the first collision information and the second collision information to analyze and determine whether the target vehicle has a risk of collision. If there is a risk of collision, based on the target dangerous driving operation and the corresponding target driving condition, combined with the target driver's actual dangerous driving operation and actual driving condition, analyze the possibility of the target vehicle colliding with the current target vehicle, and then verify whether the target vehicle does have a risk of collision, so as to more accurately evaluate the collision situation of the target vehicle. Finally, after verifying that there is a risk of collision, the target vehicle is likely to have a collision accident, so a safety warning is directly sent to the collision verification personnel. Once the target vehicle collides, the personnel can go to the scene in time to provide rescue, thereby improving the timeliness of rescue for people involved in the collision accident.

[0059] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0060] See also Figure 3 , is a schematic diagram of the structure of a vehicle collision recognition device provided in an embodiment of the present application. The vehicle collision recognition device can be implemented as all or part of the device through software, hardware or a combination of both. The device includes a data acquisition module 11, a risk determination module 12, a working condition determination module 13 and a risk warning module 14.

[0061] The data acquisition module 11 is used to acquire the recorder video and GPS data of the target vehicle before and after the current time when the target vehicle vibrates at the current time, and acquire the in-vehicle video of the target vehicle within a preset time period before the current time. The GPS data is used to characterize the location information and speed information of the target vehicle; The risk determination module 12 is used to determine the first collision information according to the GPS data, determine the recorder video as the second collision information, and determine whether there is a collision risk for the target vehicle according to the first collision information and the second collision information; The operating condition determination module 13 is used to determine, if yes, at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determine the actual driving condition of the target vehicle before the current time; The risk warning module 14 is used to verify whether the target vehicle has a risk of collision based on actual dangerous driving operations, actual driving conditions, at least one target dangerous driving operation that the target driver is prone to, and the corresponding target driving conditions. If so, a safety warning is sent to the terminal of the collision verification personnel. The target driving conditions are driving conditions that are likely to cause collision accidents when the corresponding target dangerous driving operations are present.

[0062] Optionally, the risk warning module 14 is specifically used for: Obtaining historical dangerous driving operations of the driver in the vehicle when the historical vehicle collided, counting the first occurrence number of each historical dangerous driving operation, and selecting a first number of historical dangerous driving operations from each historical dangerous driving operation in descending order of the first occurrence number as target dangerous driving operations; Obtaining historical driving conditions of the vehicle when a collision accident occurred under a single target dangerous driving operation by a historical driver, counting the second occurrence number of each historical driving condition, and selecting a second number of historical driving conditions from each historical driving condition in descending order of the second occurrence number to determine as target driving conditions for the corresponding target dangerous driving operation; Calculating a first weight for each target dangerous driving operation, and calculating a second weight for each target driving condition corresponding to each target dangerous driving operation, the first weight being a ratio of a first occurrence number of each target dangerous driving operation to a sum of first occurrence numbers of all target dangerous driving operations, and the second weight being a ratio of a second occurrence number of a single target driving condition corresponding to the target dangerous driving operation to a sum of second occurrence numbers of all corresponding target driving conditions; Based on each actual dangerous driving operation, the actual driving condition, the first weight and the corresponding second weights, it is verified whether the target vehicle has a risk of collision.

[0063] Optionally, the risk warning module 14 is specifically used for: Determine the target dangerous driving operation that has an actual driving condition in the corresponding target driving conditions as the key driving operation, and when the actual dangerous driving operation is the key driving operation, determine the corresponding actual dangerous driving operation as the reference driving operation; Calculating a first product of a first weight of each reference driving operation and a second weight of a corresponding actual driving condition, and summing the first products to obtain a sum of the first products; If the sum of the first products is greater than a preset product sum threshold, it is verified that there is a risk of collision with the target vehicle.

[0064] Optional, such as Figure 4 As shown, the device also includes an operation reminder module 15, which is specifically used to: If the target vehicle does not have a collision risk, the current driving condition of the target vehicle is determined, and the target dangerous driving operation in the current driving condition among the corresponding target driving conditions is determined as an important driving operation; Calculating a second product of the first weight of each important driving operation and the second weight of the corresponding current driving condition, and summing the second products to obtain a sum of the second products; If the sum of the second products is greater than a preset product sum threshold, then when the target driver has at least one important driving operation, the corresponding important driving operation is determined as a risky driving operation, and the important driving operations other than the risky driving operation are determined as remaining driving operations; A maximum second product is selected from the second products corresponding to the remaining driving operations, and the remaining driving operation corresponding to the maximum second product is determined as a vigilant driving operation, and a reminder message is sent to the vehicle-mounted terminal of the target vehicle for the vigilant driving operation.

[0065] Optionally, the device further includes a working condition reminder module 16, which is specifically used for: If the sum of the second products is not greater than a preset product and threshold, calculating a third product of the first weight of each risky driving operation and the second weight of each corresponding target driving condition; Summing the third products corresponding to the same target driving condition to obtain the corresponding sum of the third products, and selecting the maximum sum of the third products from the sums of the third products; When the maximum sum of the third products is greater than a preset product sum threshold, the target driving condition corresponding to the maximum sum of the third products is determined as a alert driving condition. If the alert driving condition is different from the current driving condition, a reminder message is sent to the on-board terminal of the target vehicle for the alert driving condition.

[0066] Optionally, the device further includes a cause troubleshooting module 17, which is specifically used to: Obtaining historical dangerous driving operations of the driver in the vehicle when the historical vehicle collided, counting the first occurrence number of each historical dangerous driving operation, and selecting a first number of historical dangerous driving operations from each historical dangerous driving operation in descending order of the first occurrence number as target dangerous driving operations; Obtaining historical regions where collision positions on the vehicle are located when collision accidents occur under a single target dangerous driving operation by historical drivers, counting the third occurrence times of each historical region, and selecting a third number of historical regions from each historical region in descending order of the third occurrence times to determine a target region corresponding to the target dangerous driving operation; Calculating a first weight for each target dangerous driving operation, and calculating a third weight for each target area corresponding to each target dangerous driving operation, the first weight being a ratio of a first occurrence number of each target dangerous driving operation to a sum of first occurrence numbers of all target dangerous driving operations, and the third weight being a ratio of a third occurrence number of a single target area corresponding to the target dangerous driving operation to a sum of third occurrence numbers of all corresponding target areas; During the driving process of the target vehicle, if the target driver performs a target dangerous driving operation, the corresponding target dangerous driving operation is determined as an operation to be recorded, and a mapping relationship is established between each operation to be recorded and the corresponding occurrence time node in the real-time video, where the real-time video is a driving video obtained in real time from the driving recorder in the target vehicle; When the target vehicle switches from a driving state to a normal parking state and there are unknown collision marks, the cause of the unknown collision marks is checked based on the mapping relationships, the first weights and the corresponding third weights.

[0067] Optionally, the cause troubleshooting module 17 is specifically used to: The target area where the unknown collision trace is located is determined as the key area, the target dangerous driving operation in the corresponding target area that has the key area is determined as the collision-inducing operation, the operation to be recorded in each collision-inducing operation is determined as the final operation, and the corresponding final time node is determined according to the mapping relationship of the final operation; Calculating a fourth product of the first weight of each final operation and the third weight of the corresponding focus area; According to each fourth weight, the playback order of the corresponding final time node in the real-time video is determined, and based on each playback order, the cause of the occurrence of the unknown collision trace is investigated. The larger the fourth product is, the earlier the corresponding playback order is.

[0068] It should be noted that, when executing the vehicle collision recognition method, the vehicle collision recognition device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the vehicle collision recognition device provided in the above embodiment and the vehicle collision recognition method embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0069] The embodiment of the present application further discloses a computer-readable storage medium, and the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, a vehicle collision identification method of the above embodiment is adopted.

[0070] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer-readable medium includes but is not limited to the above-mentioned components.

[0071] Among them, through this computer-readable storage medium, a vehicle collision identification method of the above embodiment is stored in a computer-readable storage medium, and is loaded and executed on a processor to facilitate the storage and application of the above method.

[0072] An embodiment of the present application further discloses an electronic device, wherein a computer program is stored in a computer-readable storage medium, and when the computer program is loaded and executed by a processor, the above-mentioned vehicle collision identification method is adopted.

[0073] The electronic device may be a desktop computer, a laptop computer, a cloud server or other electronic device, and the electronic device includes but is not limited to a processor and a memory. For example, the electronic device may also include input and output devices, a network access device, and a bus.

[0074] Among them, the processor can adopt a central processing unit (CPU). Of course, according to actual usage, 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. can also be adopted. The general-purpose processor can adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0075] Among them, the memory can be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device, or it can be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD) or a flash memory card (FC) equipped on the electronic device. Moreover, the memory can also be a combination of an internal storage unit and an external storage device of the electronic device. The memory is used to store computer programs and other programs and data required by the electronic device. The memory can also be used to temporarily store data that has been output or is to be output, and this application does not impose any restrictions on this.

[0076] Among them, through this electronic device, a vehicle collision identification method of the above embodiment is stored in the memory of the electronic device, and is loaded and executed on the processor of the electronic device, which is convenient for use.

[0077] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. This application is intended to cover any variation, use or adaptive change of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A vehicle collision identification method, characterized in that: The method comprises: When the target vehicle vibrates at the current time, the recorder video and GPS data of the target vehicle before and after the current time are obtained, and the in-vehicle video of the target vehicle within a preset time period before the current time is obtained, and the GPS data is used to characterize the location information and speed information of the target vehicle; Determine first collision information according to the GPS data, determine the recorder video as second collision information, and determine whether the target vehicle has a collision risk according to the first collision information and the second collision information; If yes, then determining at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determining the actual driving condition of the target vehicle before the current time; Based on the actual dangerous driving operations, the actual driving conditions, at least one target dangerous driving operation that is prone to occur by the target driver, and the corresponding target driving conditions, it is verified whether the target vehicle has a risk of collision. If so, a safety warning is sent to the terminal of the collision verification personnel. The target driving conditions are driving conditions that are likely to cause collision accidents when the corresponding target dangerous driving operations are present.

2. The vehicle collision recognition method according to claim 1, characterized in that: The verifying whether the target vehicle has a collision risk according to each of the actual dangerous driving operations, the actual driving conditions, at least one target dangerous driving operation that is likely to occur by the target driver, and the corresponding target driving conditions specifically includes: Acquire historical dangerous driving operations of the driver in the vehicle when the vehicle collided historically, count the first occurrence times of each of the historical dangerous driving operations, and select a first number of historical dangerous driving operations from each of the historical dangerous driving operations in descending order of the first occurrence times to determine as target dangerous driving operations; Acquire historical driving conditions of the vehicle when a collision accident occurred under a single target dangerous driving operation by a historical driver, count the second number of occurrences of each of the historical driving conditions, and select a second number of historical driving conditions from each of the historical driving conditions in descending order of the second number of occurrences to determine as target driving conditions for the corresponding target dangerous driving operation; Calculating a first weight for each of the target dangerous driving operations, and calculating a second weight for each of the target driving conditions corresponding to each of the target dangerous driving operations, wherein the first weight is a ratio of a first occurrence number of each target dangerous driving operation to a sum of first occurrence numbers of all target dangerous driving operations, and the second weight is a ratio of a second occurrence number of a single target driving condition corresponding to the target dangerous driving operation to a sum of second occurrence numbers of all corresponding target driving conditions; Based on each of the actual dangerous driving operations, the actual driving conditions, the first weights and the corresponding second weights, it is verified whether the target vehicle has a risk of collision.

3. The vehicle collision recognition method according to claim 2, characterized in that: The verifying whether the target vehicle has a collision risk based on each of the actual dangerous driving operations, the actual driving conditions, the first weights and the corresponding second weights specifically includes: Determine the target dangerous driving operation that exists in the actual driving condition in the corresponding target driving conditions as the key driving operation, and when the actual dangerous driving operation is the key driving operation, determine the corresponding actual dangerous driving operation as the reference driving operation; Calculating a first product of a first weight of each reference driving operation and a second weight of a corresponding actual driving condition, and summing the first products to obtain a sum of the first products; If the sum of the first products is greater than a preset product sum threshold, it is verified that there is a risk of collision with the target vehicle.

4. The vehicle collision recognition method according to claim 2, characterized in that: The method further comprises: If the target vehicle does not have a collision risk, determining the current driving condition of the target vehicle, and determining the target dangerous driving operation that exists in the current driving condition in each corresponding target driving condition as an important driving operation; Calculating a second product of the first weight of each of the important driving operations and the second weight of the corresponding current driving condition, and summing the second products to obtain a sum of the second products; If the sum of the second products is greater than a preset product sum threshold, when the target driver performs at least one important driving operation, the corresponding important driving operation is determined as a risky driving operation, and the important driving operations other than the risky driving operation are determined as remaining driving operations; A maximum second product is selected from the second products corresponding to the remaining driving operations, and the remaining driving operation corresponding to the maximum second product is determined as a vigilant driving operation, and a reminder message is sent to the vehicle-mounted terminal of the target vehicle for the vigilant driving operation.

5. The vehicle collision recognition method according to claim 4, characterized in that: The method further comprises: If the sum of the second products is not greater than a preset product-sum threshold, calculating a third product of the first weight of each of the risky driving operations and the second weight of each corresponding target driving condition; Summing the third products corresponding to the same target driving condition to obtain a corresponding sum of the third products, and selecting a maximum sum of the third products from the sums of the third products; When the sum of the maximum third products is greater than a preset product sum threshold, the target driving condition corresponding to the maximum sum of the third products is determined as a vigilant driving condition; if the vigilant driving condition is different from the current driving condition, a reminder message is sent to the on-board terminal of the target vehicle for the vigilant driving condition.

6. The vehicle collision recognition method according to claim 1, characterized in that: The method further comprises: Acquire historical dangerous driving operations of the driver in the vehicle when the vehicle collided historically, count the first occurrence times of each of the historical dangerous driving operations, and select a first number of historical dangerous driving operations from each of the historical dangerous driving operations in descending order of the first occurrence times to determine as target dangerous driving operations; Obtaining historical regions where collision positions on the vehicle are located when collision accidents occur under a single target dangerous driving operation by historical drivers, counting the third number of occurrences of each of the historical regions, and selecting a third number of historical regions from each of the historical regions in descending order of the third number of occurrences to determine a target region corresponding to the target dangerous driving operation; Calculating a first weight for each of the target dangerous driving operations, and calculating a third weight for each of the target areas corresponding to each of the target dangerous driving operations, wherein the first weight is a ratio of a first occurrence number of each target dangerous driving operation to a sum of first occurrence numbers of all target dangerous driving operations, and the third weight is a ratio of a third occurrence number of a single target area corresponding to the target dangerous driving operation to a sum of third occurrence numbers of all corresponding target areas; During the driving process of the target vehicle, if the target driver performs a target dangerous driving operation, the corresponding target dangerous driving operation is determined as an operation to be recorded, and a mapping relationship between each of the operations to be recorded and a corresponding occurrence time node in a real-time video is established, where the real-time video is a driving video obtained in real time from a driving recorder in the target vehicle; When the target vehicle switches from a driving state to a normal parking state and there are unknown collision marks, the cause of the unknown collision marks is checked based on the mapping relationships, the first weights and the corresponding third weights.

7. The vehicle collision recognition method according to claim 6, characterized in that: The step of investigating the cause of the unknown collision trace based on the mapping relationships, the first weights and the corresponding third weights specifically includes: Determine the target area where the unknown collision trace is located as a key area, determine the target dangerous driving operation in the key area in each corresponding target area as an induced collision operation, determine the operation to be recorded in each of the induced collision operations as a final operation, and determine the corresponding final time node according to the mapping relationship of the final operation; Calculating a fourth product of the first weight of each of the final operations and the third weight of the corresponding key area; According to each of the fourth products, the playback order of the corresponding final time nodes in the real-time video is determined, and based on each of the playback orders, the cause of the occurrence of the unknown collision traces is investigated. The larger the fourth product is, the earlier the corresponding playback order is.

8. A vehicle collision recognition device, characterized in that: include: A data acquisition module (11) is used to acquire the recorder video and GPS data of the target vehicle before and after the current time when the target vehicle vibrates at the current time, and acquire the in-vehicle video of the target vehicle within a preset time period before the current time, wherein the GPS data is used to represent the position information and speed information of the target vehicle; A risk determination module (12) is used to determine first collision information based on the GPS data, determine the recorder video as second collision information, and determine whether the target vehicle has a collision risk based on the first collision information and the second collision information; A working condition determination module (13) is used to determine, if yes, at least one actual dangerous driving operation of the target driver within a preset time period before the current time based on the in-vehicle video, and determine the actual driving working condition of the target vehicle before the current time; A risk warning module (14) is used to verify whether the target vehicle has a risk of collision based on each of the actual dangerous driving operations, the actual driving conditions, at least one target dangerous driving operation that the target driver is likely to perform, and the corresponding target driving conditions, and if so, send a safety warning to a terminal of a collision verification personnel, wherein the target driving conditions are driving conditions that are likely to cause a collision accident when the corresponding target dangerous driving operation is performed.

9. A computer-readable storage medium having a computer program stored therein, characterized in that: When the computer program is loaded and executed by a processor, the method according to any one of claims 1 to 7 is adopted.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor loads and executes the computer program, the method according to any one of claims 1 to 7 is adopted.

Images