Distracted driving detection method, computer device, storage medium, and intelligent device

Abstract

The present application relates to the technical field of vehicles, particularly provides a distracted driving detection method, a computer device, a storage medium, and an intelligent device, and aims to improve the reliability of distracted driving detection. In order to achieve this purpose, the method provided by the present application comprises: on the basis of motion information of a vehicle and motion information of out-of-vehicle targets, obtaining out-of-vehicle targets having the trajectories overlapping with the trajectory of the vehicle within a future preset time duration as candidate targets; on the basis of the candidate targets, obtaining a risk target having the risk of collision with the vehicle; obtaining a gaze state of a driver of the vehicle on the risk target; and on the basis of the gaze state, determining whether the driver is in a distracted driving state. By means of the method, false detection will not occur when there is no target or no risk target on a road, and when there are many vehicles or risk vehicles on the road, whether the driver is in the distracted driving state can be accurately determined on the basis of any of the risk targets, thereby achieving high sensitivity and reliability.

Description

Distracted driving detection method, computer equipment, storage medium and intelligent device

[0001] This application claims priority to Chinese patent application 202311684597.1 filed on December 11, 2023, with the invention name “Distracted driving detection method, computer equipment, storage medium and intelligent device”. The entire contents of the above Chinese patent application are incorporated into this application by reference. Technical Field

[0002] The present application relates to the field of vehicle technology, and in particular to a distracted driving detection method, computer equipment, storage medium, and intelligent device. Background Art

[0003] To ensure safe driving, drivers are typically monitored for distracted driving and alerted when distracted driving is detected. Current conventional detection methods primarily divide the vehicle into distraction zones (such as the interior area, rearview mirrors, etc.) and non-distraction zones (such as the front window area, left and right rearview mirrors, etc.). The system then identifies whether the driver's gaze is in the distraction zone or the non-distraction zone at each moment, calculates the percentage of time the driver's gaze is in the distraction zone over a period of time, and uses this percentage to determine whether the driver is distracted. The system also categorizes the degree of distracted driving (such as mild, moderate, or severe) based on the percentage.

[0004] However, this method detects distracted driving solely through a binary classification based on whether the driver's line of sight falls within a distracting or non-distracting area. This method is inflexible and cannot be applied to increasingly complex driving conditions. For example, in some driving conditions, there are no vehicles or no risky vehicles on the road. However, as long as the driver's line of sight meets the above conditions, it will be determined to be distracted driving and an alarm will be issued, which in turn affects the driver's safe driving. For another example, in some driving conditions, there are many vehicles or risky vehicles on the road, and timely and accurate detection of driver distraction is required to issue an alarm. However, the above-mentioned regularity is very strong, and the driver's line of sight must meet the above conditions for the alarm to be issued. This may delay the alarm, affect the vehicle's safe driving, and even cause a traffic accident.

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

[0006] Summary of the Invention

[0007] In order to overcome the above-mentioned defects, the present application is proposed to provide a distracted driving detection method, a computer device, a storage medium and an intelligent device that solve or at least partially solve the technical problem of how to improve the reliability of distracted driving detection.

[0008] In a first aspect, a distracted driving detection method is provided, the method comprising:

[0009] According to the motion information of the vehicle and the motion information of the external target, obtaining the external target whose trajectory overlaps with the vehicle within a preset time period in the future as a candidate target;

[0010] Acquire, based on the candidate targets, risk targets that may have a collision risk with the vehicle;

[0011] obtaining a gaze state of the driver of the vehicle on the risk target;

[0012] According to the gaze state, it is determined whether the driver is in a distracted driving state.

[0013] In one technical solution of the distracted driving detection method, obtaining a risk target that has a collision risk with the vehicle based on the candidate target includes:

[0014] Obtaining an overlapping moment when the candidate target and the vehicle have overlapping trajectories;

[0015] Obtaining the remaining collision time of the candidate target at the current moment based on the time between the current moment and the overlapping moment;

[0016] According to the remaining collision time, it is determined whether to take the candidate target as a risk target.

[0017] In one technical solution of the above distracted driving detection method, obtaining the remaining collision time of the candidate target at the current moment includes:

[0018] The duration between the current moment and the overlapping moment is taken as the first duration;

[0019] Obtaining a longitudinal overlap distance between the candidate target and the vehicle at the overlap moment, and a longitudinal speed difference at the overlap moment;

[0020] obtaining a ratio of the longitudinal overlapping distance to the longitudinal speed difference as a second duration;

[0021] The remaining collision time is obtained according to the difference between the first time and the second time.

[0022] In one technical solution of the distracted driving detection method, determining whether to treat the candidate target as a risk target based on the remaining collision time includes:

[0023] Determining whether the remaining collision time satisfies a first preset condition;

[0024] If so, the candidate target is taken as a risk target;

[0025] If not, the candidate target is not considered as a risk target.

[0026] In one technical solution of the distracted driving detection method, the determining whether to treat the candidate target as a risk target based on the remaining collision time further includes:

[0027] Obtaining the remaining collision time of the candidate target at multiple consecutive moments before and / or after the current moment;

[0028] Obtaining a change trend of the remaining collision time according to the remaining collision time of the candidate target at the current moment and multiple consecutive moments before and / or after the current moment;

[0029] Determining whether the remaining collision time at the current moment and a plurality of consecutive moments before and / or after the current moment all meet a first preset condition, and whether the change trend meets a second preset condition;

[0030] If so, the candidate target is taken as a risk target;

[0031] If not, the candidate target is not considered as a risk target.

[0032] In one technical solution of the above distracted driving detection method, the method further includes determining whether the remaining time to collision satisfies a first preset condition by:

[0033] Determining whether the remaining collision time is less than or equal to a preset time threshold;

[0034] If so, the remaining collision time satisfies the first preset condition;

[0035] If not, the remaining collision time does not meet the first preset condition.

[0036] In one technical solution of the above distracted driving detection method, the method further includes determining whether a changing trend of the remaining time to collision satisfies a second preset condition by:

[0037] Determining whether the change trend is a decreasing trend, and whether the difference between the remaining collision time at each two adjacent moments is greater than the time difference between each two adjacent moments;

[0038] If so, the change trend satisfies the second preset condition;

[0039] If not, the change trend does not meet the second preset condition.

[0040] In one technical solution of the above distracted driving detection method, obtaining the remaining collision time of the candidate target at multiple consecutive moments before and / or after the current moment includes:

[0041] Obtaining position information of the candidate target relative to the vehicle;

[0042] determining an error level of motion information of the candidate target based on the orientation information;

[0043] Obtaining a preset number of remaining collision times that matches the error level;

[0044] According to the number, the remaining collision time of the candidate target at a plurality of consecutive moments before and / or after the current moment is obtained.

[0045] In one technical solution of the above distracted driving detection method, determining whether the driver is in a distracted driving state based on the gaze state includes:

[0046] If the driver is not looking at the risk target, the driver is in a distracted driving state;

[0047] If the driver is looking at the risk object, the driver is not in a distracted driving state.

[0048] In one technical solution of the above distracted driving detection method, the method further includes: outputting an alarm message when it is determined that the driver is in a distracted driving state;

[0049] The alarm information includes the position information of the risk target relative to the vehicle.

[0050] In a second aspect, a computer device is provided, comprising a processor and a storage device, wherein the storage device is suitable for storing a plurality of program codes, and the program codes are suitable for being loaded and run by the processor to execute the method described in any one of the technical solutions of the above-mentioned distracted driving detection method.

[0051] In a third aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored in the computer-readable storage medium, wherein the program codes are suitable for being loaded and run by a processor to execute the method described in any one of the technical solutions of the above-mentioned distracted driving detection method.

[0052] In a fourth aspect, a smart device is provided, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program, and when the computer program is executed by the at least one processor, the method described in any one of the technical solutions of the above-mentioned distracted driving detection method is implemented.

[0053] The above one or more technical solutions of this application have at least one or more of the following beneficial effects:

[0054] In the technical solution for implementing the distracted driving detection method provided in the present application, based on the motion information of the vehicle and the motion information of external targets (such as other vehicles), external targets whose trajectories overlap with the vehicle within a preset time period in the future can be obtained as candidate targets; risk targets that pose a risk of collision with the vehicle can be obtained based on the candidate targets; the driver's gaze state on the risk targets can be obtained, and whether the driver is in a distracted driving state can be judged based on the gaze state. Based on the above embodiment, it is possible to distinguish which external targets are risk targets and which are not risk targets, and whether the driver is in a distracted driving state can be judged only based on the driver's gaze state on the risk targets, and non-risk targets will not be considered. In this way, when there are no targets or risk targets on the road, false detection will not occur as in the prior art, and the driver's safe driving of the vehicle will not be affected. When there are many vehicles on the road or many risk vehicles, based on the driver's gaze state on any risk target, it can be accurately judged whether the driver is in a distracted driving state, with high sensitivity and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The disclosure of this application will become more easily understood with reference to the accompanying drawings. Those skilled in the art will readily appreciate that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Among them:

[0056] FIG1 is a flow chart showing the main steps of a distracted driving detection method according to an embodiment of the present application;

[0057] FIG2 is a flow chart showing the main steps of a method for obtaining risk targets according to an embodiment of the present application;

[0058] FIG3 is a flowchart illustrating the main steps of a distracted driving detection method according to another embodiment of the present application;

[0059] FIG4 is a schematic diagram of the main structure of a computer device according to an embodiment of the present application.

[0060] FIG5 is a schematic diagram of the main structure of a smart device according to an embodiment of the present application.

[0061] List of reference numerals: 11: storage device; 12: processor; 21: memory; 22: processor. DETAILED DESCRIPTION

[0062] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the scope of protection of the present application.

[0063] In the description of this application, a "processor" may include hardware, software, or a combination of the two. A processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. A processor has data and / or signal processing capabilities. A processor may be implemented in software, hardware, or a combination of the two. Computer-readable storage media include any suitable medium capable of storing program code, such as a magnetic disk, a hard disk, an optical disk, flash memory, read-only memory, random access memory, and the like.

[0064] The relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, and based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.

[0065] The user personal information processed by this application will vary depending on the specific product / service scenario and is subject to the specific scenario in which the user uses the product / service. This may involve the user's account information, device information, driving information, vehicle information, or other related information. This application will treat the user's personal information and its processing with a high degree of diligence.

[0066] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that comply with industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0067] The following describes an embodiment of the distracted driving detection method provided in this application.

[0068] Referring to FIG1 , FIG1 is a flow chart illustrating the main steps of a distracted driving detection method according to an embodiment of the present application. As shown in FIG1 , the distracted driving detection method in the embodiment of the present application mainly includes the following steps S101 to S104 .

[0069] Step S101: Based on the motion information of the vehicle and the motion information of the external targets, external targets whose trajectories overlap with the vehicle within a preset time period in the future are obtained as candidate targets.

[0070] Targets outside the vehicle include but are not limited to motor vehicles, non-motor vehicles, pedestrians, cyclists, etc.

[0071] The motion information of the vehicle and the target outside the vehicle includes but is not limited to the direction, speed, position and other information of the motion. Based on this information, the motion trajectory of the vehicle and the target outside the vehicle can be predicted, and then based on the motion trajectory, it can be determined whether the vehicle and the target outside the vehicle will collide; if a collision occurs, it indicates that there is a risk of collision between the target outside the vehicle and the vehicle, and this target outside the vehicle needs to be treated as a risk target; otherwise, this target outside the vehicle does not need to be treated as a risk target.

[0072] Specifically, the system can predict the vehicle's first trajectory within a preset timeframe based on the vehicle's motion information. It can also predict the second trajectory of each external target within a preset timeframe based on the vehicle's motion information. The system then determines whether the first and second trajectories overlap, selects the overlapping second trajectories, and identifies the external targets with these second trajectories as candidate targets. If the trajectory of an external target overlaps with that of the vehicle, it indicates a potential collision at the overlapping location, and the parking target has a high collision risk.

[0073] Furthermore, when predicting motion trajectories, trajectory points at each moment within a preset future time period can be sequentially predicted and connected to form a motion trajectory. Therefore, both the first and second motion trajectories are actually composed of multiple trajectory points. When determining whether the first and second motion trajectories overlap, the trajectory points at each moment can be sequentially determined to determine whether they overlap. If any trajectory point overlap is detected, then it can be determined that the first and second motion trajectories overlap.

[0074] Step S102: Based on the candidate targets, a risk target with a collision risk with the vehicle is obtained. In this embodiment, the candidate targets can be used as risk targets, or the candidate targets can be further screened and the screened candidate targets can be used as risk targets.

[0075] Step S103: Obtain the gaze state of the vehicle driver on the risk target.

[0076] The gaze state may include gazing at a risk target and not gazing at a risk target.

[0077] In this embodiment, the position where the driver's sight is directed can be obtained. If the position is the same as the risk target, the driver is looking at the risk target; otherwise, the driver is not looking at the risk target.

[0078] Step S104: judging whether the driver is in a distracted driving state based on the gaze state. According to the gaze state, it can be judged whether the driver is paying attention to the risk target, thereby determining whether the driver is in a distracted driving state.

[0079] In some embodiments, if the driver is not looking at the risk target based on the gaze state, the driver is determined to be in a distracted driving state; if the driver is looking at the risk target based on the gaze state, the driver is determined to be not in a distracted driving state. Based on this, whether the driver is in a distracted driving state can be determined conveniently and accurately.

[0080] In some embodiments, when a driver is determined to be distracted, an alarm message is output. The alarm message may include information about the location of the risk target relative to the vehicle. This alarm message allows the driver to accurately observe the risk target while maintaining focus, allowing timely countermeasures to ensure safe driving. Furthermore, in some embodiments, the alarm message output process can monitor whether the driver begins to focus on the risk target. If so, the alarm message output may cease. Otherwise, the alarm message output may continue until the driver focuses on the risk target.

[0081] Based on the method described in steps S101 to S104 above, it is possible to distinguish between risky and non-risky objects outside the vehicle. Distracted driving is determined solely based on the driver's gaze on risky objects, without considering non-risky objects. This prevents false detections when there are no objects or risky objects on the road, potentially impacting the driver's safe driving. When the road is crowded or high in risky vehicles, distracted driving can be accurately determined based on any risky object, demonstrating high sensitivity and reliability.

[0082] In an application scenario according to an embodiment of the present application, a device capable of executing the distracted driving detection method of an embodiment of the present application can be configured on a vehicle. While the user is driving the vehicle, this device can monitor in real time whether the driver is driving distracted. If the driver is driving distracted, an alarm message will be output to remind the driver until the driver concentrates and stops driving distracted, thereby ensuring the safety of the vehicle.

[0083] The above step S102 is further explained below.

[0084] In some implementations of the above step S102 , the following steps S201 to S203 shown in FIG. 2 may be performed to obtain risk targets based on candidate targets.

[0085] Step S201: Obtain the overlapping moment when the candidate target and the vehicle have overlapping trajectories.

[0086] As described in the aforementioned step S101, when determining whether the first and second motion trajectories overlap, the overlap of the trajectory points at each moment can be determined sequentially. The moment at which the overlapping trajectory point occurs can be obtained as the overlap moment. Furthermore, to ensure vehicle safety, the moment at which the candidate target and vehicle trajectories first overlap can be obtained, and subsequent steps can be executed based on this first overlap moment.

[0087] Step S202: According to the time between the current moment and the overlapping moment, the remaining collision time of the candidate target at the current moment is obtained.

[0088] In this embodiment, the above duration may be directly used as the remaining collision time, or the duration may be reduced and the reduced duration may be used as the remaining collision time.

[0089] Step S203: Determine whether to designate the candidate target as a risk target based on the remaining collision time. A longer remaining collision time indicates more time left for the driver to operate the vehicle and avoid a collision, and the risk of collision between the candidate target and the vehicle is lower. Conversely, a shorter remaining collision time indicates a higher risk of collision between the candidate target and the vehicle. Therefore, the remaining collision time can be used to determine the risk of collision between the candidate target and the vehicle, and further determine whether to designate the candidate target as a risk target. For example, if the risk is high, the candidate target is designated as a risk target; otherwise, it is not designated as a risk target.

[0090] Based on the method described in steps S201 to S203 above, after obtaining the candidate target, the risk level can be judged according to the remaining collision time of the candidate target, thereby accurately obtaining the risk target.

[0091] The above steps S202 and S203 are further explained below.

[0092] (1) Further explanation of step S202.

[0093] In some implementations of the above step S202 , the remaining collision time of the candidate target at the current moment may be obtained through the following steps S2021 to S2024 .

[0094] Step S2021: The duration between the current moment and the overlapping moment is taken as the first duration.

[0095] Step S2022: Obtain the longitudinal overlap distance between the candidate target and the vehicle at the overlap moment, and the longitudinal speed difference between the candidate target and the vehicle at the overlap moment.

[0096] Longitudinal direction refers to the forward direction of travel of the vehicle.

[0097] As described in the aforementioned step S201, when determining whether the first and second motion trajectories overlap, the moment at which the overlapping trajectory point occurs can be obtained as the overlap moment. Because the first and second motion trajectories are composed of multiple discrete trajectory points, each two trajectory points are separated by a period of time, such as 0.1 seconds. Therefore, it is possible that the motion trajectories of the vehicle and the external object have already overlapped before the aforementioned overlapping trajectory point. By the time the vehicle and the external object reach the aforementioned overlapping trajectory point, there will be a longitudinal overlap distance between them. This overlap distance is what is obtained in this step.

[0098] The longitudinal velocity difference is the difference between the longitudinal velocity of the candidate target at the moment of overlap and the longitudinal velocity of the vehicle at the moment of overlap.

[0099] Step S2023: Obtain the ratio of the longitudinal overlap distance to the longitudinal speed difference as the second duration. The candidate target's longitudinal speed may be greater than or less than the vehicle's longitudinal speed; therefore, the longitudinal speed difference may be positive or negative. After obtaining the ratio of the longitudinal overlap distance to the longitudinal speed difference, the absolute value of this ratio may be obtained as the second duration.

[0100] Step S2024: Obtain the remaining collision time according to the difference between the first time and the second time.

[0101] As described in step S2022, the trajectories of the vehicle and the external target may have overlapped before the overlapping trajectory point. Therefore, the overlap moment obtained based on the overlapping trajectory point cannot accurately represent the moment of collision between the vehicle and the external target. The longitudinal overlap distance, however, can represent the distance between the vehicle and the external target that continued to collide after the collision moment. Therefore, the second duration calculated based on the longitudinal overlap distance can represent the duration of the ongoing collision. Subtracting the second duration from the first duration accurately determines the duration from the current moment to the moment of collision between the vehicle and the external target, i.e., the remaining collision duration.

[0102] Based on the method described in steps S2021 to S2024 above, a more accurate remaining collision time can be obtained, thereby improving the accuracy of the risk target and ensuring vehicle driving safety.

[0103] (2) Further explanation of step S203.

[0104] In some implementations of the above step S203, whether to treat the candidate target as a risk target can be determined through the following steps 11 to 13.

[0105] Step 11: Determine whether the remaining collision time meets the first preset condition;

[0106] If satisfied, go to step 12; if not satisfied, go to step 13.

[0107] In some embodiments, it is possible to determine whether the remaining collision time is less than or equal to a preset time threshold. If so, the remaining collision time is determined to satisfy a first preset condition, i.e., the remaining collision time is relatively short. If not, the remaining collision time is determined to not satisfy the first preset condition, i.e., the remaining collision time is relatively long. Those skilled in the art may flexibly set the value of the preset time threshold based on actual needs, and this embodiment does not impose specific limitations on this. For example, in some preferred embodiments, the preset time threshold may be 3 seconds.

[0108] Step 12: Select candidate targets as risk targets.

[0109] Step 13: Do not consider candidate targets as risky targets.

[0110] Based on the methods described in steps 11 to 13 above, risk targets can be quickly and accurately screened out by conditionally matching the remaining collision time.

[0111] In other implementations of the above step S203, it is also possible to determine whether to treat the candidate target as a risk target through the following steps 21 to 25.

[0112] Step 21: Obtain the remaining collision time of the candidate target at multiple consecutive moments before and / or after the current moment.

[0113] The aforementioned step S202 describes a method for obtaining the remaining collision time for a candidate target at the current moment. The same method can be used to obtain the remaining collision time for moments before and after the current moment. This embodiment stores the remaining collision time for each candidate target at a given moment. When the user needs the remaining collision time, it can be directly accessed.

[0114] Step 22: Obtain a change trend of the remaining collision time based on the remaining collision time of the candidate target at the current moment and multiple consecutive moments before and / or after the current moment.

[0115] The change trend refers to the change trend of the remaining collision time of the candidate target during the time period formed at the above moment. For example, the change trend may be gradually decreasing, gradually increasing, or irregular fluctuations.

[0116] Step 23: Determine whether the remaining collision time at the current moment and a plurality of consecutive moments before and / or after the current moment all meet a first preset condition and whether the change trend meets a second preset condition.

[0117] If both the first and second preset conditions are met, go to step 24;

[0118] If the first and second preset conditions cannot be met simultaneously, go to step 25.

[0119] The method for determining whether the first preset condition is met is the same as the related methods in the aforementioned steps 11 to 13, and will not be repeated here.

[0120] If the remaining time to collision at each of the current moment and multiple consecutive moments before and / or after the current moment satisfies the first preset condition, then the remaining time to collision at each moment is relatively short. If the changing trend of the remaining time to collision satisfies the second preset condition, then the changing trend indicates a high collision risk between the candidate target and the vehicle.

[0121] As can be seen from the description of the aforementioned embodiments, obtaining the remaining collision time requires the use of motion information from the vehicle and external targets. This motion information is primarily acquired by sensors and other devices on the vehicle. For example, the vehicle's IMU acquires vehicle speed, and the vehicle's camera captures external images, which then performs target recognition on the external images to obtain information such as the position and speed of external targets. Because sensors and other devices are required to acquire motion information, this motion information may also contain errors due to errors in these devices at certain moments. The remaining collision time at a particular moment, derived based on this motion information, may be biased. Therefore, if the remaining collision time at a single moment is used alone to determine whether a candidate target is a risky target, misjudgment may occur. To address this issue, this embodiment further incorporates the changing trend of the remaining collision time to make a judgment, addressing this limitation and preventing misjudgments. If the remaining collision time at each moment satisfies the first pre-set condition, and the changing trend of the remaining collision time also satisfies the second pre-set condition, then the candidate target can be determined to have a very high collision risk and be considered a risky target.

[0122] Step 24: Select candidate targets as risk targets.

[0123] Step 25: Do not consider candidate targets as risky targets.

[0124] Based on the methods described in steps 21 to 25 above, the value of the remaining collision time at each moment and the changing trend of the remaining collision time over a period of time can be combined to accurately determine whether the candidate target is a risky target and prevent misjudgment.

[0125] The above steps 21 and 23 are further explained below.

[0126] 1. Explain step 21.

[0127] In some implementations of step 21 , the remaining collision time of the candidate target at a plurality of consecutive moments before and / or after the current moment may be obtained through the following steps 211 to 214 .

[0128] Step 211: Obtain the position information of the candidate target relative to the vehicle.

[0129] The position information may include the direction of motion and position of the candidate target relative to the vehicle.

[0130] The direction of movement may include same direction, opposite direction, lateral direction, etc., wherein same direction refers to moving in the same direction as the vehicle, opposite direction refers to moving in the opposite direction of the vehicle, and lateral direction refers to moving laterally in front of the vehicle.

[0131] The location may include being in the same lane as the vehicle, being in an adjacent lane to the vehicle, etc.

[0132] Step 212: Determine the error level of the motion information of the candidate target based on the orientation information.

[0133] As can be seen from the description of step 23 above, motion information is primarily acquired by sensors and other devices on the vehicle. For external targets at different orientations, the motion information obtained using these devices may have different errors. In this embodiment, a correspondence between different orientation information and different error levels can be pre-set. After the orientation information is obtained in step 211, this correspondence is matched to obtain the corresponding error level.

[0134] In some embodiments, four types of position information may be set, namely:

[0135] 1. The external object is in the same lane as the vehicle and is traveling in front of the vehicle in the same direction as the vehicle.

[0136] 2. The external object is in the adjacent lane of the vehicle and is traveling in the same direction as the vehicle in front of the vehicle.

[0137] 3. The external object is in the vehicle's lane or an adjacent lane, but traveling in the opposite direction of the vehicle.

[0138] 4. The external object is in the adjacent lane of the vehicle and is moving laterally in front of the vehicle.

[0139] Among the above four types of position information, the error level corresponding to the first type is the smallest, the error level corresponding to the second type is the second, and the error levels corresponding to the third and fourth types are the largest.

[0140] Step 213: Obtain a preset number of remaining collision times that matches the error level. The higher the error level, the more remaining collision times need to be obtained to ensure an accurate trend in the remaining collision time. Based on this, a predefined relationship between different error levels and the number of remaining collision times can be pre-set. After the error level is obtained in step 212, this relationship is then matched to obtain the number of remaining collision times.

[0141] Taking the implementation in the aforementioned step 212 as an example, the smallest, second largest, and largest error levels can respectively obtain the remaining collision time at 5 moments, 10 moments, and 20 moments.

[0142] Step 214: Obtain the remaining collision time of the candidate target at multiple consecutive moments before and / or after the current moment based on the quantity.

[0143] Taking the example of obtaining the remaining collision time at 10 moments, in addition to the current moment, you can also obtain the remaining collision time at 9 consecutive moments after the current moment, or you can obtain the remaining collision time at 9 consecutive moments before the current moment, or you can obtain the remaining collision time at multiple consecutive moments before and after the current moment at the same time, as long as it is guaranteed to be 9 moments.

[0144] Based on the method described in steps 211 to 214 above, a certain number of remaining collision times can be obtained in a targeted manner based on the position information of the candidate target relative to the vehicle, so as to ensure that the changing trend obtained based on these remaining collision times can accurately determine whether there is a high collision risk between the candidate target and the vehicle.

[0145] 2. Explain step 23.

[0146] In some implementations of step 23 , it may be determined through the following steps 231 to 233 whether the changing trend of the remaining collision time meets the second preset condition.

[0147] Step 231: Determine whether the change trend is a decreasing trend, and whether the difference between the remaining collision time at each two adjacent moments is greater than the time difference between each two adjacent moments.

[0148] If the change trend is decreasing and the difference is greater than the time difference, it indicates that the vehicle is rapidly approaching the candidate target and the risk of collision with the candidate target is relatively high. In this case, it can be determined that the second preset condition is met, and the process goes to step 232. Otherwise, the process goes to step 233.

[0149] Step 232: Determine whether the change trend meets the second preset condition.

[0150] Step 233: Determine whether the change trend does not meet the second preset condition.

[0151] Based on the method described in steps 231 to 233 above, the accuracy and reliability of risk target judgment based on the second preset condition can be improved.

[0152] The distracted driving detection method of the embodiment of the present application will be briefly described again with reference to FIG3. As shown in FIG3, distracted driving detection can be performed by the following steps.

[0153] First, the vehicle motion parameters of the current vehicle and the target motion parameters of the target outside the vehicle are obtained, and the trajectory is calculated based on the motion parameters of the two to obtain the motion trajectory of the current vehicle (i.e., the vehicle trajectory in Figure 3) and the motion trajectory of the target outside the vehicle (i.e., the target trajectory in Figure 3).

[0154] The vehicle motion parameters may include the vehicle's position, longitudinal / lateral velocity, longitudinal / lateral acceleration, and vehicle size information (including at least the vehicle's length, width, and height). The target motion parameters may include the target's position, longitudinal velocity, longitudinal acceleration, yaw rate, and target size information (including at least the target's length, width, and height). When performing trajectory extrapolation, the vehicle's three-dimensional position at preset intervals (such as 0.1 seconds) within a preset time period (such as 5 seconds) in the future can be obtained as trajectory points, and these trajectory points are connected to form the vehicle's trajectory. Similarly, the target trajectory can also be obtained using the same method.

[0155] After obtaining the own vehicle trajectory and the target trajectory, the remaining collision time ttc (time to crash) of each external target at the current moment is calculated. The calculation method can adopt the relevant method in the above method embodiment.

[0156] After obtaining the remaining collision time (TTC) of an external target at the current moment, a determination is made as to whether the TTC satisfies a condition. The condition can be the first pre-set condition in the aforementioned method embodiment, or the determination method can employ the relevant method in the aforementioned method embodiment. If the condition is satisfied, the external target is considered a risk target, and its attributes are updated to risk attributes. If the condition is not satisfied, the external target is considered a non-risk target, and its attributes are updated to non-risk attributes. The TTC of the external target at the current moment is stored as a frame of multi-frame risk information for the target.

[0157] Then, for the risk attributed external target, a multi-frame risk assessment is performed. Specifically, the remaining collision time of the external target at the current moment and multiple consecutive moments before and / or after the current moment can be obtained, and a determination can be made as to whether the changing trend of these remaining collision time periods meets the second preset condition. This determination can be made using the relevant methods described in the aforementioned method embodiments.

[0158] If the conditions are met, the driver's attention state to the target outside the vehicle is queried (that is, the gaze state in the aforementioned embodiment), and whether there is a multi-frame risk is determined based on the attention state, that is, the changing trend of the remaining collision time indicates a collision risk; if there is a risk, an alarm is issued; if there is no risk, the multi-frame risk level is cleared, that is, the information on the degree of collision risk in the changing trend of the remaining collision time is cleared, that is, the changing trend of the remaining collision time is set to no collision risk.

[0159] If the conditions are not met, continue with the target multi-frame risk judgment.

[0160] The above is a brief description of the distracted driving detection method shown in FIG3 .

[0161] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of the present application, different steps do not have to be performed in such an order. They can be performed simultaneously (in parallel) or in other orders. These adjusted solutions are equivalent to the technical solutions described in this application, and therefore will also fall within the scope of protection of this application.

[0162] It will be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment of the present application can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium can include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium that can carry the computer program code. It should be noted that the content contained in the computer-readable storage 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 storage media do not include electric carrier signals and telecommunication signals.

[0163] Another aspect of the present application provides a computer device.

[0164] In one embodiment of a computer device according to the present application, the computer device primarily includes a storage device and a processor. The storage device may be configured to store a program for executing the distracted driving detection method of the aforementioned method embodiment, and the processor may be configured to execute the program stored in the storage device, including but not limited to a program for executing the distracted driving detection method of the aforementioned method embodiment. Referring to FIG. 4 , FIG. 4 exemplarily illustrates a storage device 11 and a processor 12 communicatively connected via a bus. For ease of illustration, only the portions relevant to this embodiment are shown. For specific technical details not disclosed, please refer to the method section of the present embodiment.

[0165] In some possible implementations, a computer device may include multiple storage devices and multiple processors. The multiple processors may be processors deployed on the same device. For example, the computer device may be a high-performance device composed of multiple processors, and the multiple processors may be processors configured on the high-performance device. Furthermore, the multiple processors may be processors deployed on different devices. For example, the computer device may be a server cluster, and the multiple processors may be processors on different servers in the server cluster.

[0166] Another aspect of the present application provides a computer-readable storage medium.

[0167] In one embodiment of a computer-readable storage medium according to the present application, the computer-readable storage medium may be configured to store a program for executing the distracted driving detection method of the aforementioned method embodiment. This program may be loaded and executed by a processor to implement the aforementioned distracted driving detection method. For ease of illustration, only the portions relevant to the present embodiment are shown. For specific technical details not disclosed, please refer to the method section of the present embodiment. The computer-readable storage medium may be a storage device formed of various electronic devices. Optionally, in the embodiments of the present application, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0168] Another aspect of the present application provides a smart device.

[0169] In an embodiment of an intelligent device according to the present application, the intelligent device may include at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program, and when executed by the at least one processor, the computer program implements the method described in any of the above embodiments. The intelligent device described in this application may include a driving device, a smart car, a robot, and other devices. Referring to FIG. 5 , FIG. 5 exemplarily illustrates a memory 21 and a processor 22 communicatively connected via a bus.

[0170] In some embodiments of the present application, the smart device may further include at least one sensor for sensing information. The sensor is communicatively connected to any of the types of processors mentioned in this application. Optionally, the smart device may further include an autonomous driving system for guiding the smart device to drive autonomously or provide assisted driving. The processor communicates with the sensor and / or autonomous driving system to perform the method described in any of the above embodiments.

[0171] Thus far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A distracted driving detection method, characterized in that: The method comprises: According to the motion information of the vehicle and the motion information of the external target, obtaining the external target whose trajectory overlaps with the vehicle within a preset time period in the future as a candidate target; Acquire, based on the candidate targets, risk targets that have a collision risk with the vehicle; Acquiring a gaze state of the driver of the vehicle on the risk target; According to the gaze state, it is determined whether the driver is in a distracted driving state.

2. The method according to claim 1, characterized in that The step of acquiring, based on the candidate targets, a risk target that has a collision risk with the vehicle includes: Obtaining an overlapping moment when the trajectory of the candidate target overlaps with that of the vehicle; According to the time between the current moment and the overlapping moment, obtaining the remaining collision time of the candidate target at the current moment; According to the remaining collision time, it is determined whether to take the candidate target as a risk target.

3. The method according to claim 2, characterized in that The obtaining of the remaining collision time of the candidate target at the current moment includes: The duration between the current moment and the overlapping moment is taken as the first duration; Acquire a longitudinal overlap distance between the candidate target and the vehicle at the overlap moment, and a longitudinal speed difference at the overlap moment; Acquire a ratio of the longitudinal overlapping distance to the longitudinal speed difference as a second duration; The remaining collision time is obtained according to the difference between the first time and the second time.

4. The method according to claim 2, characterized in that: The determining, according to the remaining collision time, whether to take the candidate target as a risk target includes: Determining whether the remaining collision time satisfies a first preset condition; If yes, the candidate target is taken as a risk target; If not, the candidate target is not considered as a risk target.

5. The method according to claim 2, characterized in that: The determining, according to the remaining collision time, whether to take the candidate target as a risk target further includes: Obtaining the remaining collision time of the candidate target at multiple consecutive moments before and / or after the current moment; According to the remaining collision time of the candidate target at the current moment and multiple consecutive moments before and / or after the current moment, obtaining a change trend of the remaining collision time; Determining whether the remaining collision time at the current moment and a plurality of consecutive moments before and / or after the current moment all meet a first preset condition, and whether the change trend meets a second preset condition; If yes, the candidate target is taken as a risk target; If not, the candidate target is not considered as a risk target.

6. The method according to claim 4 or 5, characterized in that: The method further includes determining whether the remaining collision time meets a first preset condition by: Determining whether the remaining collision time is less than or equal to a preset time threshold; If so, the remaining collision time satisfies the first preset condition; If not, the remaining collision time does not meet the first preset condition.

7. The method according to claim 5, characterized in that The method further includes determining whether the change trend of the remaining collision time meets the second preset condition by: Determine whether the change trend is a decreasing trend, and whether the difference between the remaining collision time of each two adjacent moments is greater than the time difference between each two adjacent moments; If yes, then the change trend satisfies the second preset condition; If not, the change trend does not meet the second preset condition.

8. The method according to claim 5, characterized in that The obtaining of the remaining collision time of the candidate target at a plurality of consecutive moments before and / or after the current moment includes: Acquiring position information of the candidate target relative to the vehicle; Determining an error level of motion information of the candidate target according to the orientation information; Obtaining a preset number of remaining collision times that matches the error level; According to the number, the remaining collision time of the candidate target at a plurality of consecutive moments before and / or after the current moment is obtained.

9. The method according to claim 1, characterized in that: The step of judging whether the driver is in a distracted driving state according to the gaze state includes: If the driver does not look at the risk target, the driver is in a distracted driving state; If the driver is looking at the risk object, the driver is not in a distracted driving state.

10. The method according to claim 1 or 9, characterized in that: The method further includes: outputting warning information when it is determined that the driver is in a distracted driving state; Wherein, the alarm information includes the position information of the risk target relative to the vehicle.

11. A computer device comprising a processor and a storage device, wherein the storage device is suitable for storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and executed by the processor to execute the distracted driving detection method according to any one of claims 1 to 10.

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

13. A smart device, characterized in that: include: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the distracted driving detection method according to any one of claims 1 to 10 is implemented.

Images