Vehicle and method of monitoring thereof

Abstract

The application discloses a vehicle and a monitoring method thereof. The monitoring method comprises the following steps: acquiring an image collected from a scene where the vehicle is located; triggering an alarm state of the vehicle in the case that a first target object is determined to stay within a first distance range of the vehicle and a staying time of the first target object reaches a first time length, the first target object being capable of being determined through image recognition, and the first target object having a human body feature. According to the monitoring method, the first target object is determined to stay at a relatively short distance from the vehicle for a relatively long time and has a human body feature, so it can be considered that a person stays around the vehicle for a long time and may contact the vehicle, and the vehicle has a potential risk, and thus the alarm state is triggered to facilitate further monitoring and reduce the number of missed alarms.

Description

Technical Field

[0001] This invention relates to the field of vehicle monitoring technology, and in particular to a vehicle and its monitoring method. Background Technology

[0002] In related technologies, vehicles can have parking monitoring functions. When the user leaves the vehicle, the parking monitoring function can be activated to detect whether the vehicle has been scratched or collided in the off-duty scenario. However, current parking monitoring functions often only trigger alarms when the vehicle is scratched or collided, making it difficult to detect situations that only cause minor impacts to the vehicle or have potential risks, thus easily leading to missed alarms. Summary of the Invention

[0003] The present invention provides a vehicle and a method for monitoring the same to solve at least one of the aforementioned technical problems.

[0004] A vehicle monitoring method of the present invention includes:

[0005] Acquire images of the scene where the vehicle is located;

[0006] If a first target object is determined to be within a first distance range of the vehicle and the duration of the stay reaches a first duration, the vehicle's alarm state is triggered. The first target object can be identified by image recognition and has human characteristics.

[0007] The aforementioned monitoring method determines that if the first target object stays at a relatively close distance to the vehicle for a long time and has human characteristics, it can be considered that someone has been staying around the vehicle for a long time and may come into contact with the vehicle, posing a potential risk to the vehicle. This triggers an alarm state, which facilitates further monitoring and reduces the number of missed alarms.

[0008] In one optional technical solution of the present invention, the monitoring method includes:

[0009] When the first target object first enters the first distance range of the vehicle, the duration of the first target object's stay is timed.

[0010] During the timing process, if the duration during which the first target object is not continuously identified reaches a second duration, and / or the number of consecutive image frames in which the first target object is not identified reaches a first frame number, the timing is stopped.

[0011] During the timing process, if the duration during which the first target object is not continuously identified is less than the second duration, and / or the number of consecutive image frames in which the first target object is not identified is less than the first number of frames, the timing is maintained. This avoids misidentification due to dropped image frames.

[0012] In one optional technical solution of the present invention, the monitoring method includes:

[0013] If the first target object leaves the first distance range, determine whether the first target object is still within a second distance range of the vehicle, where the second distance range is greater than or equal to the first distance range;

[0014] If it is determined that no first target object is present within a second distance range of the vehicle, and this condition is maintained for a third duration, the vehicle's alarm status is deactivated. This improves accuracy in determining the absence of a first target object around the vehicle.

[0015] In one optional technical solution of the present invention, the monitoring method includes:

[0016] When the first target object leaves the first distance range for the first time, start timing the duration during which the first target object does not remain within the second distance range;

[0017] During the timing process, if the duration of continuous identification of the first target object reaches the fourth duration, and / or the number of consecutive image frames in which the first target object is identified reaches the second frame number, the timing is stopped.

[0018] During the timing process, if the duration during which the first target object is not continuously identified is less than the fourth duration, and / or the number of consecutive image frames in which the first target object is not identified is less than the second number of frames, the timing is maintained. This avoids misidentification due to dropped image frames.

[0019] In one optional technical solution of the present invention, the monitoring method includes:

[0020] Obtain object information obtained by radar scanning of the scene where the vehicle is located;

[0021] If a second target object is detected entering the vehicle's third distance range, and the second target object remains within the third distance range for a duration of five hours, the vehicle's alarm is triggered. The second target object can be identified through object information, and the third distance range is smaller than the first distance range. This achieves effective monitoring even when an object is approaching the vehicle at close range.

[0022] In one optional technical solution of the present invention, the monitoring method includes:

[0023] When the second target object first enters the third distance range, the duration of the second target object being within the third distance range is started.

[0024] During the timing process, if the duration of the second target object not being identified reaches the sixth duration, and / or the number of consecutive image frames in which the second target object is not identified reaches the third frame number, the timing shall be stopped.

[0025] During the timing process, if the duration during which the second target object is not continuously identified is less than the sixth duration, and / or the number of consecutive image frames in which the second target object is not identified is less than the third number of frames, the timing is maintained. This avoids misidentification due to dropped image frames.

[0026] In one optional technical solution of the present invention, the monitoring method includes:

[0027] If the second target object leaves the third distance range, determine whether the second target object is still within the fourth distance range of the vehicle, where the fourth distance range is greater than or equal to the third distance range;

[0028] If it is determined that there is no second target object within the fourth distance range and the duration reaches the seventh time, the alarm status of the vehicle is deactivated.

[0029] And / or, the monitoring method includes:

[0030] If the distance fluctuation between the second target and the vehicle is less than a preset fluctuation distance, and this fluctuation persists for a duration of nine hours, the vehicle's alarm status will be deactivated. This ensures that the second target does not pose a potential risk to the vehicle.

[0031] In one optional technical solution of the present invention, the monitoring method includes:

[0032] When the second target object first leaves the third distance range, the timer starts to count the duration during which the second target object is not present in the fourth distance range;

[0033] During the timing process, if the duration of the second target object not being identified reaches the eighth duration, and / or the number of consecutive image frames in which the second target object is not identified reaches the fourth frame number, the timing shall be stopped.

[0034] During the timing process, if the duration for which the second target object is not continuously identified is less than the eighth duration, and / or the number of consecutive image frames in which the second target object is not identified is less than the fourth number of frames, the timing is maintained. This avoids misidentification due to dropped image frames.

[0035] In one optional technical solution of the present invention, the monitoring method includes:

[0036] If the vehicle's alarm is triggered, video recording will be initiated of the scene where the vehicle is located. This allows for timely recording of the specific process of a collision or scrape involving the vehicle.

[0037] A vehicle according to the present invention includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the monitoring method described in any of the above optional technical solutions.

[0038] The aforementioned vehicle is identified by determining that the first target object stays at a relatively close distance to the vehicle for a relatively long time and has human characteristics. This indicates that someone has been lingering around the vehicle for an extended period and may come into contact with it, posing a potential risk to the vehicle. This triggers an alarm state to facilitate further monitoring and reduce the number of missed alarms.

[0039] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0040] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0041] Figure 1 This is a partial structural schematic diagram of the vehicle according to an embodiment of the present invention;

[0042] Figure 2 This is a schematic diagram of a scenario where a vehicle is being monitored according to an embodiment of the present invention;

[0043] Figure 3 This is a flowchart illustrating the monitoring method according to an embodiment of the present invention;

[0044] Figure 4 This is a schematic diagram of a vehicle according to an embodiment of the present invention;

[0045] Figure 5 This is a schematic diagram of another module of the vehicle according to an embodiment of the present invention.

[0046] Explanation of key component symbols:

[0047] 100 vehicles;

[0048] Vehicle body 11, vehicle end 12, camera 13, lidar 14;

[0049] Control module 111, image sensor 112, radar sensor 113;

[0050] Memory 201, Processor 202. Detailed Implementation

[0051] In the description of this invention, some of the disclosed content has been shown accordingly in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The following description with reference to the accompanying drawings is exemplary and is only used to explain the invention, and should not be construed as limiting the invention.

[0052] In the description of this invention, many different contents or examples are disclosed to implement different structures of the invention. To simplify the disclosure of this invention, the components and arrangements of specific examples are described below. Of course, these are merely examples and are not intended to limit the invention.

[0053] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0054] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0055] Furthermore, the present invention may repeat reference numbers and / or reference letters in different examples. Such repetition is for the purpose of simplification and clarity and does not in itself indicate the relationship between the various situations and / or settings discussed.

[0056] In related technologies, vehicles can have parking monitoring functions. When the user leaves the vehicle, the parking monitoring can be activated to monitor the surrounding environment in the absence of the vehicle, thereby promptly identifying any scratches, collisions, or other potential risks to the vehicle and issuing an alarm.

[0057] However, the vehicle's parking monitoring function may be limited by the settings of the detection equipment, resulting in inconsistent vibration detection sensitivity for different parts of the vehicle, and it is easily affected by other scene factors (such as rain, environmental noise, etc.), which may lead to missed alarms and false alarms.

[0058] Furthermore, in scenarios where the vehicle is out of its sight, users cannot be informed of the vehicle's status in a timely manner, and if an alarm is triggered, the user will not be notified immediately. Although some related technologies can notify users of a collision through appropriate notification strategies, they cannot display nearby images, leaving users unaware of the specific circumstances of the collision and potentially causing anxiety.

[0059] Based on the above, the technical solution of the present invention can provide a parking monitoring strategy, which may include triggering identification and reminders for parking monitoring, recording of images, and remote prompts and timely notifications to users, thereby improving the accuracy of parking monitoring, helping users to record core events that pose a risk to the vehicle and retain evidence for further reporting, while also enabling timely access to vehicle information, reducing user anxiety, and improving the user experience.

[0060] Specifically, the present invention can provide a corresponding method and vehicle to implement and apply the aforementioned parking monitoring strategy.

[0061] Please refer to Figure 1 In this invention, the vehicle 100 may include a vehicle body 11 and a vehicle terminal 12. A passenger compartment may be formed inside the vehicle body 11. The vehicle terminal 12 may be disposed within the passenger compartment. The vehicle terminal 12 may serve as the control assembly of the vehicle 100, thereby controlling other controls of the vehicle 100 (such as doors, windows, and air conditioning). In some cases, the vehicle terminal 12 may have a large display screen (not shown). The vehicle terminal 12 can display relevant information to the user through the large display screen, such as information about the status of the vehicle 100, surrounding environmental information, navigation information, multimedia information, etc. The user can determine the required information through the large display screen and then operate the vehicle terminal 12, thereby achieving the effect of controlling the vehicle 100 through the vehicle terminal 12.

[0062] Vehicle 100 can communicate with corresponding terminal devices. Specifically, vehicle terminal 12 may include a communication unit (not shown). Users can establish a communication channel with the communication unit through the terminal device, enabling vehicle terminal 12 to communicate with the terminal device via the communication unit. Users can issue corresponding commands by operating the terminal device, which can then send the commands to the communication unit, allowing vehicle terminal 12 to receive the commands and execute the corresponding operations. Terminal devices may include digital keys, smartphones, tablets, servers, personal computers, and wearable smart devices. Communication methods between vehicle 100 and terminal devices may include wireless networks, mobile communication networks (3G, 4G, 5G, etc.), Bluetooth, infrared, etc.

[0063] Vehicle 100 can be powered on and off. When powered on, vehicle 100 can supply power to the corresponding controls. When powered off, vehicle 100 can de-power the corresponding controls. Vehicle 100 can be powered on when unlocking and powered off when locking.

[0064] Vehicle 100 may have a sentry mode. When vehicle 100 is in sentry mode, it can monitor its surroundings and issue an alert if a collision, scrape, or other potential risk is detected. Users can activate sentry mode by operating vehicle terminal 12 or by operating the terminal device.

[0065] In some cases, vehicle 100 can start alarm monitoring by powering down after activating sentry mode via vehicle terminal 12, or it can activate sentry mode to start alarm monitoring after powering down based on the monitoring command received from the terminal device.

[0066] Among them, for the above-mentioned alarm monitoring activation method, a first delay period can be set, that is, the vehicle 100 can be powered down after the sentry mode is activated through the vehicle terminal 12, and alarm monitoring will be activated after the first delay period, or the sentry mode can be activated according to the monitoring command sent by the terminal device after power-down, and alarm monitoring will be activated after the first delay period.

[0067] It's understandable that the need for Vehicle 100 to activate Sentry Mode primarily arises when the user is away from the vehicle. In such cases, the user might be preparing to leave but still near Vehicle 100, or they might be performing actions outside the vehicle. This means the user could actively detect and perceive a collision or scrape involving Vehicle 100. In these situations, alarm monitoring is less valuable and may even trigger a false alarm based on the user's actions. By setting a first delay, alarm monitoring will only activate after the user has completely left the vehicle, thus improving the effective utilization of Vehicle 100's battery power and memory.

[0068] The first delay duration can be adjusted according to specific circumstances or calibrated through actual testing. The value range of the first delay duration is [8, 12] (unit: seconds). In some cases, the first delay duration can be 10 seconds.

[0069] In some cases, when vehicle 100 is in sentry mode, vehicle 100 can detect the user's corresponding actions and thus pause or stop alarm monitoring.

[0070] Specifically, the user's corresponding operations can include controlling vehicle 100 for remote parking, operating the trunk, and remotely unlocking vehicle 100. The user can control vehicle 100 to activate sentry mode, and then control vehicle 100 to enter remote parking mode to perform remote parking in the current parking scenario. User operations on the trunk can include opening or closing the trunk, and can also include retrieving or placing items from the trunk while it is open.

[0071] Since the user's actions indicate that they are operating vehicle 100 or are near vehicle 100, and vehicle 100 determines whether a collision has occurred by monitoring vibrations, this avoids misinterpreting vibrations caused by the user's actions as a collision. In other words, by eliminating scenarios involving user interaction, false alarms caused by user actions can be avoided.

[0072] In some cases, when alarm monitoring is paused or stopped, vehicle 100 can maintain the paused or stopped alarm monitoring in response to the corresponding instruction or request. If the corresponding instruction or request is canceled, or if the opposite instruction or request is received, vehicle 100 can restart alarm monitoring after a set restart time.

[0073] Additionally, when vehicle 100 is in sentry mode, if it receives a command to exit sentry mode, it will generate and report corresponding monitoring information upon exiting sentry mode. This monitoring information may include relevant information about the surrounding environment monitored by vehicle 100 while in sentry mode, such as monitoring duration, monitored events, and corresponding time points. Users can determine the situation monitored by vehicle 100 while in sentry mode based on this monitoring information.

[0074] In some cases, when vehicle 100 is performing alarm monitoring, the user can remotely view vehicle 100. Specifically, when vehicle 100 is performing alarm monitoring through visual recognition, the user can obtain the image obtained by visual recognition during alarm monitoring. The image can be sent from vehicle terminal 12 to terminal device, allowing the user to view the image on the terminal device.

[0075] Please combine Figure 2 The vehicle 100 may include a camera 13. The camera 13 may be installed inside the vehicle body 11 and may face one side of the vehicle 100 to capture images. The vehicle 100 may have multiple cameras 13 arranged around the vehicle body 11. The multiple cameras 13 may each face different directions to capture images, thereby enabling a surround view of the environment around the vehicle 100 and the detection of collisions or scrapes from different directions.

[0076] For vehicle 100, sentry mode and remote viewing can have different priorities. Remote viewing can have a higher priority than sentry mode. When using remote viewing, the user can access any of the cameras 13 on vehicle 100 used to monitor the surrounding environment, thereby viewing the condition of vehicle 100 from any direction.

[0077] Specifically, in some cases, when vehicle 100 is in sentry mode, if it receives a remote viewing request from a user, it will pause sentry mode to allow the user to remotely view the camera 13. Upon receiving a user's instruction to stop remote viewing, vehicle 100 can resume sentry mode to continue alarm monitoring. In some cases, vehicle 100 can record video of the surrounding environment in sentry mode to preserve evidence. During video recording, if vehicle 100 receives a remote viewing request from a user, it will pause video recording and allow the user to remotely view the camera 13.

[0078] In other words, vehicle 100 will prioritize responding to user actions to inform the user of its surroundings. By pausing sentry mode and / or video recording, while vehicle 100 transmits images to the user for remote viewing, it avoids consuming excessive memory resources that would otherwise be required for visual recognition of the surrounding environment under alarm monitoring. This ensures sufficient memory resources are available to prioritize uploading image data to the terminal device, allowing the user to promptly grasp the situation around vehicle 100. Moreover, when the user is remotely viewing, they can confirm the surrounding environment themselves, eliminating the need for vehicle 100 to perform active alarm monitoring.

[0079] For Vehicle 100, events can be classified into risk levels based on their varying degrees of risk. In some cases, events can be categorized into Level 1 and Level 2 risk events. A Level 1 risk event indicates a potential risk but may not necessarily cause substantial damage, while a Level 2 risk event indicates a high probability of an accident. Vehicle 100 can monitor for Level 1 risk events using visual recognition and for Level 2 risk events using vibration recognition.

[0080] If vehicle 100 is identified as having a risk event of a corresponding level through alarm monitoring, appropriate measures can be taken. In some cases, when a Level 1 risk event is detected, it can be recorded and saved as an image or video as evidence if the user discovers a scratch on vehicle 100. When a Level 2 risk event is detected, a corresponding reminder message can be sent to the user when saving the relevant image or video, informing the user that they need to proactively check vehicle 100 for damage.

[0081] If an accident is detected during vehicle 100's alarm monitoring, an alarm will be triggered and the vehicle will enter alarm mode. In alarm mode, vehicle 100 can record relevant information to preserve evidence based on the risk level of the event, and / or send alert messages to the user.

[0082] The alarm state can be maintained for a second delay, during which vehicle 100 can continuously acquire images and / or video to record a sufficient amount of the accident process, and / or continuously send reminder messages to the user to prevent the user from missing messages. After the second delay, vehicle 100 can automatically exit the alarm state and then re-enter alarm monitoring.

[0083] In such a situation, if someone lingers close to vehicle 100 for an extended period, they may come into contact with it. However, whether or not this causes harm is unknown, meaning there is a potential risk. Therefore, this situation can be classified as a Level 1 risk event. The resulting damage could include scratches caused by malicious vandalism or accidental scrapes.

[0084] Vehicle 100 can detect whether anyone is staying within a first set distance around vehicle 100 via camera 13. If the duration of the stay is greater than or equal to the first set duration, a level one risk event is determined to have occurred, and an alarm will be triggered.

[0085] Specifically, vehicle 100 can first detect whether there are people around. When it detects someone entering a first set distance for the first time, it will start timing the duration of that person's stay. Figure 2 In this context, the first predetermined distance can be denoted as L1. The vehicle 100 can form a first distance range with its own center as the center and the first predetermined distance as the range. The area within the first distance range can be denoted as S1.

[0086] During the timing process, if the duration of no human detection continues to the second preset duration, and / or the number of consecutive image frames in the acquired images that do not detect a human reaches the first preset frame number, it indicates that the person has left vehicle 100, and the timing will stop. If the duration of no human detection is less than the second preset duration, and / or the number of consecutive image frames in the acquired images that do not detect a human is less than the first preset frame number, it indicates that the person is still around vehicle 100, and the timing will continue, triggering an alarm state after the first preset duration.

[0087] Vehicle 100 can continuously acquire images at certain time intervals, thereby making the second set duration related to the first set number of frames. In some cases, vehicle 100 can continuously acquire 30 consecutive images within 1 second.

[0088] In addition, the first set distance, first set duration, second set duration, and first set frame count can all be adjusted according to specific circumstances and can all be calibrated through actual testing. The first set distance can range from [40, 60] (unit: centimeters). The first set duration can range from [8, 12] (unit: seconds). The second set duration can range from [0.8, 1.2] (unit: seconds). The first set frame count can range from [28, 32]. In some cases, the first set distance can be 50 centimeters, the first set duration can be 10 seconds, the second set duration can be 1 second, and the first set frame count can be 30.

[0089] If an alarm is triggered due to the detection of someone approaching the vehicle, the alarm will only be deactivated after the corresponding conditions are confirmed.

[0090] Specifically, vehicle 100 can use camera 13 to detect whether anyone is staying within a second preset distance around vehicle 100. When it is determined that no one is staying for a duration greater than or equal to a third preset duration, the alarm state is deactivated.

[0091] Vehicle 100 can first detect whether there are people around. When it detects that there are no people within a second set distance from vehicle 100 for the first time, it will start timing the duration of no one staying.

[0092] During the timing process, if the duration of a person being detected lasts for the fourth set duration, and / or the number of consecutive image frames in the collected images that detect a person reaches the second set number of frames, it indicates that someone is approaching vehicle 100, and the timing will stop while the alarm status remains active. If the duration of a person being detected is less than the fourth set duration, and / or the number of consecutive image frames in the collected images that detect a person is less than the second set number of frames, it indicates that no one is approaching vehicle 100, and the timing will continue until the third set duration is reached before the alarm status is deactivated.

[0093] Vehicle 100 can continuously acquire images at certain time intervals, thus making the fourth set duration related to the second set frame number. In some cases, vehicle 100 can continuously acquire 30 consecutive frames of images within 1 second.

[0094] In addition, the second set distance, third set duration, fourth set duration, and second set frame count can all be adjusted according to specific circumstances and can be calibrated through actual testing. The range of the second set distance is [40, 60] (unit: centimeters). The range of the third set duration is [8, 12] (unit: seconds). The range of the fourth set duration is [0.8, 1.2] (unit: seconds). The range of the second set frame count is [28, 32]. In some cases, the second set distance can be 50 centimeters, the third set duration can be 10 seconds, the fourth set duration can be 1 second, and the second set frame count can be 30.

[0095] Furthermore, in some cases, the second preset distance can be greater than or equal to the first preset distance. This ensures that the process of preparing to de-alarm will only begin if the distance between the person and vehicle 100 is at least the first preset distance. Conversely, the third preset duration can be greater than or equal to the first preset duration. This ensures that the process of preparing to de-alarm will only begin if the person has been sufficiently far from vehicle 100 for at least the first preset duration.

[0096] Based on the above, vehicle 100 can acquire images of its surrounding environment via camera 13 and perform human recognition on the images, thereby detecting whether anyone is approaching vehicle 100. Furthermore, the above description focuses on whether someone is approaching vehicle 100; it can be understood that it is also applicable to detecting whether obstacles other than people are approaching vehicle 100.

[0097] In such a scenario, if someone lingers close to vehicle 100 for an extended period, minor collisions to vehicle 100 may not be detected visually. In this case, radar scanning can supplement the monitoring. When a person or object is detected lingering close to vehicle 100, this can be classified as a Level 1 risk event, allowing the recording of minor scrapes to vehicle 100 that do not trigger a vibration alarm. Objects can include obstacles in the scene, as well as moving objects around vehicle 100 (such as other vehicles).

[0098] Vehicle 100 can use radar to detect whether a person or object is approaching at close range within a third preset distance around vehicle 100. When it is determined that a person or object is approaching vehicle 100 from outside the third preset distance and enters the third preset distance range of vehicle 100, and the duration of the close approach is greater than or equal to a fifth preset duration, a level one risk event is determined to have occurred, and an alarm state will be triggered.

[0099] Specifically, vehicle 100 can first detect whether there are people or objects in the surrounding area. When a person or object is detected entering the third predetermined distance for the first time, it will start timing the duration of the person or object's stay. Figure 2 In this context, the third preset distance can be represented as L2. The vehicle 100 can form a third distance range with its own center as the center and the third preset distance as the range. The area within the third distance range can be represented as S2.

[0100] During the timing process, if no person or object is detected within the third set distance of vehicle 100 for a duration that continues to the sixth set duration, or if the number of consecutive image frames in the acquired images in which no person or object is detected reaches the third set number of frames, it indicates that the person or object has left vehicle 100, and the timing will stop. If no person or object is detected within the third set distance of vehicle 100 for a duration that is less than the sixth set duration, or if the number of consecutive image frames in the acquired images in which no person or object is detected is less than the third set number of frames, it indicates that the person or object is still close to vehicle 100, and the timing will continue, triggering an alarm state after the fifth set duration.

[0101] Vehicle 100 can continuously acquire images at certain time intervals, thus making the sixth set duration related to the third set frame number. In some cases, vehicle 100 can continuously acquire 30 consecutive frames of images within 1 second.

[0102] In addition, the third set distance, fifth set duration, sixth set duration, and third set frame count can all be adjusted according to specific circumstances and can all be calibrated through actual testing. The value range of the third set distance is [28, 32] (unit: centimeters). The value range of the fifth set duration is [8, 12] (unit: seconds). The value range of the sixth set duration is [0.8, 1.2] (unit: seconds). The value range of the third set frame count is [28, 32]. In some cases, the third set distance can be 30 centimeters, the fifth set duration can be 10 seconds, the sixth set duration can be 1 second, and the third set frame count can be 30.

[0103] In addition, please combine Figure 2In some cases, the third preset distance may be less than the first preset distance. That is, within the range between the first and third preset distances, the camera 13 can capture images and perform image recognition to detect if anyone is approaching. Within the third preset distance, radar scanning and obstacle recognition can be used to detect if anyone or any object is approaching. It is understandable that at closer distances, it may be difficult to capture images that accurately identify objects. In this case, radar scanning can be used to determine the distance of the object relative to the vehicle 100, thereby determining the risk of a minor collision for the vehicle 100. Thus, by combining visual recognition and radar scanning, it is possible to monitor whether objects around the vehicle 100 are approaching it.

[0104] In cases where an alarm is triggered due to the detection of a person or object approaching the vehicle at close range, the alarm will only be deactivated after the corresponding conditions are confirmed.

[0105] Specifically, vehicle 100 can further detect whether there are people or objects remaining within a fourth preset distance around vehicle 100 via radar scanning. When it is determined that no one has remained for a duration greater than or equal to a seventh preset duration, the alarm state is deactivated.

[0106] Vehicle 100 can first detect whether there are people or objects around. When no one or object is detected within a fourth set distance from vehicle 100 for the first time, it will start timing the duration of the absence of people or objects.

[0107] During the timing process, if the duration of detecting a person or object continues to the eighth preset duration, or if the number of consecutive image frames in the captured images that detect a person or object reaches the fourth preset frame number, it indicates that a person or object is still close to the vehicle 100, and the timing will stop while maintaining the alarm status. If the duration of detecting a person or object is less than the eighth preset duration, or if the number of consecutive image frames in the captured images that detect a person or object is less than the fourth preset frame number, it indicates that the person or object has moved away from the vehicle 100, and the timing will continue until the seventh preset duration is reached before the alarm status is deactivated.

[0108] In addition, vehicle 100 can also monitor the distance change of a person or object relative to vehicle 100. If the fluctuation range of the distance change is less than the fifth set distance and is maintained for the ninth set duration, it can be indicated that an object approaching vehicle 100 at close range will not contact vehicle 100, and the alarm state can be deactivated.

[0109] Vehicle 100 can continuously acquire images at certain time intervals, thus making the eighth set duration related to the fourth set frame number. In some cases, vehicle 100 can continuously acquire 30 consecutive frames of images within 1 second.

[0110] In addition, the fourth, fifth, seventh, eighth, and ninth time settings, as well as the fourth frame rate setting, can all be adjusted according to specific circumstances and calibrated through actual testing. The fourth distance setting can range from [28, 32] (cm). The fifth distance setting can range from [4.8, 5.2] (cm). The seventh time setting can range from [2.8, 3.2] (seconds). The eighth time setting can range from [0.8, 1.2] (seconds). The ninth time setting can range from [50, 70] (seconds). The fourth frame rate setting can range from [28, 32]. In some cases, the fourth distance setting can be 30 cm, the fifth distance setting can be 5 cm, the seventh time setting can be 3 seconds, the eighth time setting can be 1 second, the ninth time setting can be 60 seconds, and the fourth frame rate setting can be 30.

[0111] Based on the above, vehicle 100 can detect the distance of objects in the surrounding environment relative to vehicle 100 through radar scanning, thereby monitoring whether there are people or objects approaching vehicle 100 at close range.

[0112] In such a situation, if vehicle 100 experiences vibration, and the vibration is sudden, irregular, and highly discrete, it is determined that a collision may occur, and thus such a situation can be classified as a level two risk event.

[0113] Specifically, vehicle 100 can monitor the vibrations it receives and generate corresponding vibration data. When the magnitude of the vibration data is greater than or equal to a first preset threshold, an alarm will be triggered, indicating that vehicle 100 has experienced a certain degree of collision or abnormal vibration. The vibration data can be the intensity of the vibrations received by vehicle 100.

[0114] When vehicle 100 is in sentry mode, it can intermittently detect offset data values ​​at corresponding time intervals, thereby obtaining multiple offset data values ​​in sequence. Vehicle 100 can process the obtained multiple offset data values ​​to obtain corresponding vibration data, and then determine whether the vibration experienced by vehicle 100 is caused by a collision between vehicle 100 and an object close to vehicle 100.

[0115] The offset data value can be detected by the gyroscope of vehicle 100. Specifically, when vehicle 100 is subjected to vibration, the gyroscope will move accordingly and generate an offset distance, which will be used as the offset data value.

[0116] Vehicle 100 can acquire the offset distance generated by the gyroscope at ten predetermined time intervals as an offset data value. Each offset data value can have a time point at which it is acquired. After the acquired offset data values ​​reach a predetermined number for the first time, if a new offset data value is acquired, the average of the previous few offset data values ​​of the new offset data value will be processed to obtain an average offset data value. Then, the squared difference between the new offset data value and the average offset data value will be processed to generate vibration data. The time point corresponding to this vibration data is the time point when the aforementioned new offset data value was acquired.

[0117] Based on the above, during the alarm monitoring process, when a new offset data value is acquired, vibration data corresponding to the new offset data value can be generated based on the new offset data value and the previous few offset data values, thereby continuously monitoring the degree of vibration experienced by the vehicle 100.

[0118] The number of offset data values ​​used to obtain the average offset data value can be set to ensure that there are enough prior samples for comparison, so as to shield against continuous and regular vibrations (such as vehicle 100 being continuously impacted by rain in a rainy weather scenario).

[0119] Both vibration data and offset data values ​​can have direction in a spatial coordinate system. The spatial coordinate system can be preset according to the vehicle 100 to form three axes (XYZ). The direction of the vibration data and offset data values ​​can be along one of the corresponding coordinate axes of the XYZ axes. The magnitude of the vibration data is greater than or equal to a first preset threshold, which can be defined as the magnitude of the vibration data along its direction being greater than or equal to the first preset threshold.

[0120] In addition, the first set threshold, the set quantity, and the tenth set duration can all be adjusted according to specific circumstances and can all be calibrated through actual testing. The set quantity can range from [8, 12]. The tenth set duration can range from [18, 22] (unit: milliseconds). In some cases, the set quantity can be 10, and the tenth set duration can be 20 milliseconds.

[0121] Furthermore, in some cases, the first set threshold can be determined based on the false alarm rate and false negative rate of vibration in the test scenario. Specifically, when setting the first set threshold, the test can be conducted on the premise that the false alarm rate does not exceed 5% and the false negative rate does not exceed 5%.

[0122] In the event that an alarm is triggered due to abnormal vibration detected in vehicle 100, the alarm will only be deactivated after the corresponding conditions are confirmed to be met.

[0123] Specifically, vehicle 100 can continue to receive vibration data, and then determine whether the magnitude of the received vibration data is less than a first set threshold. If it is determined that the magnitude of the continuously received vibration data is less than the first set threshold and the duration of maintenance reaches an eleventh set duration, it can be indicated that vehicle 100 is no longer subjected to abnormal vibration, thereby determining the alarm cancellation state.

[0124] In addition, the eleventh setting duration can be adjusted according to specific circumstances, or it can be calibrated through actual testing. The value range of the eleventh setting duration is [8, 12] (unit: milliseconds). In some cases, the eleventh setting duration can be 10 milliseconds.

[0125] Based on the above, vehicle 100 can determine that an object has caused a certain degree of collision to vehicle 100 by acquiring one of the vibration data, which can ensure the sensitivity of abnormal vibration detection, and can determine that vehicle 100 is no longer being collided by acquiring multiple continuous vibration data, which can ensure accurate recording of abnormal vibration.

[0126] In such a situation, when a person or other object (such as another vehicle) approaches vehicle 100, a minor collision may occur upon contact. This minor collision may be difficult to detect because the impact is very small or the location of the impact is in a part of vehicle 100 with low vibration detection sensitivity. However, a minor collision still constitutes an abnormal collision with vehicle 100 and such a situation needs to be classified as a level two risk event.

[0127] Specifically, when a person or object is detected approaching vehicle 100, it can be determined whether the magnitude of the generated vibration data is greater than or equal to a second preset threshold. If the magnitude of the vibration data is greater than or equal to the second preset threshold, an alarm state will be triggered, indicating that vehicle 100 has experienced a minor collision. The second preset threshold is lower than the first preset threshold, which corresponds to increased sensitivity of vehicle 100 to vibration detection, enabling vehicle 100 to detect minor collisions. Vehicle 100 can use visual recognition and / or radar scanning to detect whether a person or object is approaching vehicle 100. Correspondingly, the magnitude of the vibration data being greater than or equal to the second preset threshold can be defined as the magnitude of the vibration data along its direction being greater than or equal to the second preset threshold.

[0128] Among them, vehicle 100 can detect whether it has been subjected to a minor collision by setting up an IMU (Inertial Measurement Unit).

[0129] Furthermore, the second set threshold can be adjusted according to specific circumstances or calibrated through actual testing. In some cases, the second set threshold can be determined based on the false alarm rate and false negative rate of vibration in the test scenario. Specifically, when setting the second set threshold, testing can be conducted with the premise that the false alarm rate does not exceed 8% and the false negative rate does not exceed 2%.

[0130] If a minor collision is detected and triggers an alarm, the alarm will only be deactivated after certain conditions are met.

[0131] Specifically, if a person or object is detected to be moving away from vehicle 100, and the continuous vibration data is less than the second set threshold and the duration of this condition reaches the twelfth set duration, it can be indicated that vehicle 100 is no longer subject to minor collisions, thereby determining the alarm cancellation state.

[0132] In addition, the twelfth setting duration can be adjusted according to specific circumstances, or it can be calibrated through actual testing. The value range of the twelfth setting duration is [8, 12] (unit: milliseconds). In some cases, the twelfth setting duration can be 10 milliseconds.

[0133] Based on the above, vehicle 100 can combine visual recognition and / or radar scanning to monitor whether a person or object is approaching vehicle 100. By acquiring one of the vibration data, it can determine that a person or object has caused a minor collision to vehicle 100, which can ensure the sensitivity of minor collision detection. Furthermore, by acquiring multiple continuous vibration data, it can determine that vehicle 100 is no longer being collided with, which can ensure accurate recording of minor collisions.

[0134] When an alarm is triggered, the risk level of the event can be recorded and determined. However, in some scenarios, multiple events may occur within a certain period. In such cases, multiple events can be combined and marked as triggering a single alarm.

[0135] Specifically, if an alarm state is triggered by a preceding event and has not yet been cleared, and another event occurs, the alarm state triggered by the preceding event will not be interrupted and re-entered; instead, the current alarm state will be maintained. Furthermore, since each event has its own alarm state clearing condition, the current alarm state will only be completely cleared after all events have cleared their respective conditions. In other words, all events in this process are considered to have triggered an alarm state once. This avoids frequent interruptions and restarts of alarm states. During video recording, it also avoids frequent interruptions and re-recordings that would generate multiple videos; instead, it generates a single video recording the entire process corresponding to this alarm state, allowing users to easily view the complete process of multiple events.

[0136] Specifically, if multiple events include only Level 1 risk events, the alert status and corresponding recorded video will be recorded as Level 1; if multiple events include only Level 2 risk events, the alert status and corresponding recorded video will be recorded as Level 2; if multiple events include both Level 1 and Level 2 risk events, the alert status and corresponding recorded video will be recorded as Level 2. In other words, the alert status and recorded video will be recorded according to the level of the event with the highest risk among multiple events.

[0137] Specifically, in one scenario, if a Level 1 risk event triggers an alarm, and then a Level 2 risk event occurs, the alarm status and video recording will be maintained, but recorded as Level 2 instead of Level 1. In another scenario, if a Level 2 risk event triggers an alarm, and then a Level 1 risk event occurs, the alarm status and video recording will be maintained, but recorded as Level 2 instead of Level 1.

[0138] Generally, the recording duration corresponds to the duration of the alarm state, thus ensuring a complete record of the events occurring during the alarm state. An alarm state can have a maximum duration, preventing invalid video recordings due to misidentification and avoiding unnecessary recordings caused by repeated events continuously triggering the alarm state.

[0139] In some cases, when vehicle 100 monitors the environment around vehicle 100 through visual recognition, there may be cases of misidentification or missed identification due to special scenarios, which requires additional processing for these special scenarios.

[0140] In some scenarios, stationary objects may exist around vehicle 100. Due to factors such as light and shadow interference and the shape of the object, vehicle 100 may easily misidentify the stationary object as a human body. In this case, after the object first triggers the alarm state of vehicle 100 and is recorded by video for a certain period of time, the system can determine whether the object has remained unchanged for an extended period during the next identification. If there is no change, the alarm will not be triggered again for that object, and the object will be filtered out in the object recognition model used for object identification.

[0141] Specifically, when the alarm is first triggered, video recording of the object can be performed for a set duration up to the thirteenth time. After recording is completed, when the same object is detected again, it can be determined whether the object has changed. The change in the object can be determined by a change in the distance between the object and the vehicle 100, or by a change in the object's posture relative to the vehicle 100.

[0142] In some scenarios, multiple people may walk around vehicle 100. While no one may actually linger for long, vehicle 100 continuously detects approaching people, potentially triggering an alarm and causing false alarms. In such cases, vehicle 100 can monitor the duration of one person's stay around it. If the stay is relatively short, it can be determined that no one has lingered around vehicle 100 for an extended period. Conversely, if the stay is sufficiently long, it can be determined that someone has lingered around vehicle 100 for an extended period.

[0143] Specifically, vehicle 100 can acquire images from different directions for monitoring. Each direction can correspond to a view. Under a view, the duration of time a person stays after passing by vehicle 100 can be recorded. If the duration of the same person recorded under a view is greater than or equal to the fourteenth preset duration, it can be determined that someone has stayed for an extended period of time, thus avoiding false identification.

[0144] When a person is continuously moving and is located in different positions of vehicle 100, the person can be identified in the view corresponding to the previous position, and then in the view corresponding to the next position. Thus, by corresponding to different views, it can be determined that the person has moved from one position of vehicle 100 to another position of vehicle 100. During this process, the duration of the person's stay will be continuously recorded, and the time will not be reset when the person moves to a different position of vehicle 100. This can avoid both false identification and missed identification.

[0145] In some scenarios, someone may move very close to vehicle 100. Due to the arrangement of cameras 13, it may be difficult to identify the human features of people very close to vehicle 100, potentially resulting in the inability to identify the person and causing missed identification. In such cases, the process of a person approaching vehicle 100 can be monitored to determine if someone is too close to vehicle 100.

[0146] It is understandable that a person needs to move to approach vehicle 100. This is a process of moving from far to near relative to vehicle 100, allowing vehicle 100 to detect the person's human characteristics before they get very close. If the vehicle determines that it cannot detect human characteristics, or that the human characteristics have disappeared, it can be determined that the person is too close to vehicle 100, and an alarm will be triggered.

[0147] Specifically, vehicle 100 can acquire images of its surroundings through visual recognition and identify human bodies located at or beyond a sixth preset distance from vehicle 100 based on these images. If a human body is located in the middle area of ​​the image, and then moves closer to vehicle 100 to a distance less than the sixth preset distance, the human body will not be detected in the acquired image at this point. The detectable human features will disappear in the middle area of ​​the image, thus triggering an alarm on vehicle 100. The sixth preset distance can range from [8, 12] (unit: centimeters). A sixth preset distance can be 10 centimeters.

[0148] Among them, human body features can include the overall outline of a person, or the outline of the head, torso, or limbs.

[0149] In addition, when a person is within a first set distance from vehicle 100, even if they are not too close to vehicle 100, it will be considered a level one risk event and trigger an alarm.

[0150] The central region of an image is defined relative to its edge regions. The edge regions are the outermost parts of the image furthest from the center, while the central region is the innermost part closer to the center. When human features disappear from the edge regions of the image, it corresponds to a scene where a person passes by vehicle 100. Since the person is not too close to vehicle 100, the alarm will not be triggered.

[0151] Based on the above, the technical concept of this invention can also be embodied in the following content.

[0152] Please refer to Figure 1 and Figure 3 The vehicle 100 monitoring method of the present invention may include:

[0153] 011: Acquire images of the scene where vehicle 100 is located;

[0154] 012: If it is determined that a first target object is staying within a first distance range of the vehicle 100 and the staying time reaches a first duration, the alarm state of the vehicle 100 is triggered. The first target object can be identified by image recognition and the first target object has human characteristics.

[0155] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The vehicle 100 may include a control module 111. The control module 111 may be used to: acquire images of the scene where the vehicle 100 is located; and trigger an alarm state of the vehicle 100 when it is determined that a first target object is staying within a first distance range of the vehicle 100 and the staying time reaches a first duration. The first target object can be identified by image recognition and has human characteristics.

[0156] The aforementioned monitoring method and vehicle 100, by determining that the first target object stays at a relatively close distance to vehicle 100 for a long time and has human characteristics, can be considered as someone staying around vehicle 100 for a long time and possibly coming into contact with vehicle 100, posing a potential risk to vehicle 100, thereby triggering an alarm state to facilitate further monitoring and reduce the number of missed alarms.

[0157] Specifically, in some cases, the first distance range of vehicle 100 can be a circular area centered on vehicle 100 with a first predetermined distance as its radius. The first duration can be a first predetermined duration. Vehicle 100 can perform human body recognition based on the acquired images. When human features are detected in the target object, the target object can be identified as the first target object.

[0158] Understandably, aside from the user of vehicle 100, other people generally won't intentionally linger near vehicle 100 for extended periods. By monitoring whether anyone is staying around vehicle 100 for a long time, it can be assumed that vehicle 100 poses a potential risk, such as malicious scratching or scraping. This can trigger an alarm in a timely manner. In alarm mode, a message can be sent to the user, or video recording can be initiated to preserve evidence.

[0159] Among them, Figure 4In this vehicle, 100 may also include an image sensor 112. The vehicle 100 can acquire images of the scene in which it is located using the image sensor 112. The image sensor 112 may include a camera 13. The vehicle 100 may be equipped with multiple image sensors 112. Each image sensor 112 may be configured to face one of the directions of the vehicle 100 to acquire images of the scene in that direction.

[0160] In this invention, the monitoring method may include:

[0161] When the first target object first enters the first distance range of the vehicle 100, the duration of the first target object's stay is timed.

[0162] During the timing process, if the duration during which the first target object is not continuously identified reaches the second duration, and / or the number of consecutive image frames in which the first target object is not identified reaches the first frame number, the timing shall be stopped.

[0163] During the timing process, if the duration of the continuous failure to identify the first target object is less than the second duration, and / or the number of consecutive image frames in which the first target object is not identified is less than the number of the first frames, the timing is maintained.

[0164] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: start timing the dwell time of the first target object when it first enters the first distance range of the vehicle 100; stop timing when it is determined that the duration of the first target object not being identified reaches a second duration, and / or the number of consecutive image frames of the first target object not being identified reaches a first frame number; and maintain timing when it is determined that the duration of the first target object not being identified is less than the second duration, and / or the number of consecutive image frames of the first target object not being identified is less than the first frame number. This avoids misidentification due to dropped image frames.

[0165] It is understandable that in practical applications, when identifying the first target object and determining its dwell time through acquired images, some frames may be lost due to factors such as interference from external signals. This can easily lead to misjudgment of the dwell time of the first target object, resulting in false alarms or missed alarms. By setting a second duration and the number of first frames, misjudgments caused by image frame loss can be avoided.

[0166] Specifically, in some cases, the second duration can be a second preset duration. The first frame number can be a first preset frame number.

[0167] In this invention, the monitoring method may include:

[0168] If the first target object leaves the first distance range, determine whether the first target object is still within the second distance range of vehicle 100, where the second distance range is greater than or equal to the first distance range;

[0169] If no first target object is found within the second distance range of vehicle 100, and the duration of this condition reaches the third duration, the alarm status of vehicle 100 is deactivated.

[0170] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: determine whether the first target object remains within a second distance range of the vehicle 100 when the first target object leaves the first distance range, wherein the second distance range is greater than or equal to the first distance range; and deactivate the alarm state of the vehicle 100 if it is determined that no first target object remains within the second distance range of the vehicle 100 and the duration reaches a third duration. This improves the accuracy of determining that no first target object exists around the vehicle 100.

[0171] It is understandable that once the first target object has completely moved away from vehicle 100, there is no longer any potential risk posed by the first target object, and therefore the alarm does not need to be maintained. By setting a second distance range and a third duration, it is possible to determine when the first target object has completely moved away from vehicle 100, thus facilitating the confirmation of when to deactivate the alarm.

[0172] Specifically, in some cases, the second distance range of vehicle 100 can be a circular area centered on vehicle 100 and with a second predetermined distance as the radius. The third duration can be a third predetermined duration.

[0173] In this invention, the monitoring method may include:

[0174] When the first target object leaves the first distance range for the first time, the timer starts to count the duration during which the first target object does not remain within the second distance range;

[0175] During the timing process, if the duration of continuous identification of the first target object reaches the fourth duration, and / or the number of consecutive image frames in which the first target object is identified reaches the second frame number, the timing shall be stopped.

[0176] If, during the timing process, the duration during which the first target object is not continuously identified is less than the fourth duration, and / or the number of consecutive image frames in which the first target object is not identified is less than the second number of frames, the timing is maintained.

[0177] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: start timing the duration during which no first target object remains within a second distance range when the first target object first leaves the first distance range; stop timing when, during the timing process, the duration during which the first target object is continuously identified reaches a fourth duration, and / or the number of consecutive image frames in which the first target object is identified reaches a second frame count; and continue timing when, during the timing process, the duration during which the first target object is not continuously identified is less than a fourth duration, and / or the number of consecutive image frames in which the first target object is not identified is less than a second frame count. This avoids misidentification due to dropped image frames.

[0178] It is understandable that in practical applications, when identifying the first target object and determining the duration of its departure from vehicle 100 based on acquired images, some image frames may be lost due to factors such as interference from external signals. This can easily lead to misjudgments of the duration of the first target object's departure from vehicle 100, resulting in false alarms or missed alarms. By setting a fourth duration and a second number of frames, misjudgments caused by image frame loss can be avoided.

[0179] Specifically, in some cases, the fourth duration can be a fourth preset duration. The second frame number can be a second preset frame number.

[0180] In this invention, the monitoring method may include:

[0181] Acquire object information obtained by radar scanning of the scene where vehicle 100 is located;

[0182] If a second target is determined to have entered the third distance range of vehicle 100, and the second target has been within the third distance range for a duration of five hours, the alarm state of vehicle 100 is triggered. The second target can be identified through object information, and the third distance range is smaller than the first distance range.

[0183] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: acquire object information obtained by radar scanning of the scene where vehicle 100 is located; and trigger an alarm state of vehicle 100 when it is determined that a second target has entered the third distance range of vehicle 100, and the second target has been within the third distance range for a duration of five hours. The second target can be identified through object information, and the third distance range is smaller than the first distance range. In this way, a monitoring effect can be achieved in scenarios where an object is approaching vehicle 100 at close range.

[0184] If a second target moves to within a third distance range of vehicle 100, it may be too close to be identified by image recognition. However, the distance between the second target and vehicle 100 can be determined using object information obtained from radar scanning. If, based on the distance between the second target and vehicle 100, the second target remains within the third distance range for a duration of five hours, it can be considered that the second target may pose a collision risk to vehicle 100 or present a potential danger, thus triggering an alarm. Ultimately, this ensures effective monitoring of the area around vehicle 100 when image recognition is insufficient to identify objects at close range.

[0185] Specifically, in some cases, the third distance range of vehicle 100 can be a circular area centered on vehicle 100 with a third predetermined distance as its radius. The fifth duration can be a fifth predetermined duration. Vehicle 100 can perform distance detection based on the acquired object information. When a target object is detected moving into the third distance range, the target object can be identified as the second target object. The second target object can be the first target object, or it can be any other object besides the first target object (such as another vehicle).

[0186] Among them, Figure 4 In this vehicle, 100 may also include a radar sensor 113. The vehicle 100 can use the radar sensor 113 to detect objects in the space surrounding it to obtain object information. The radar sensor 113 may include a lidar sensor 14. The vehicle 100 may be equipped with multiple radar sensors 113. Each radar sensor 113 may be configured to face one of the directions of the vehicle 100 to detect objects in the scene at that direction.

[0187] In addition, Figure 4 Based on this, when identifying objects within a first distance range using the image sensor 112 and objects within a third distance range using the radar sensor 113, the third distance range is set to be smaller than the first distance range. This allows the radar sensor 113 to detect objects that are closer to the vehicle 100. This reduces the detection range of the radar sensor 113 and the amount of detection data that needs to be processed, and improves the accuracy of distance detection for objects at close range.

[0188] In this invention, the monitoring method may include:

[0189] When the second target first enters the third distance range, start timing the duration of the second target's stay within the third distance range;

[0190] During the timing process, if the duration of the second target object not being identified reaches the sixth duration, and / or the number of consecutive image frames without the second target object being identified reaches the third frame number, the timing shall be stopped.

[0191] During the timing process, if the duration of the second target not being identified is less than the sixth duration, and / or the number of consecutive image frames in which the second target is not identified is less than the third frame number, the timing is maintained.

[0192] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: start timing the duration of the second target object being within the third distance range when the second target object first enters the third distance range; stop timing when, during the timing process, the duration of the second target object not being identified reaches the sixth duration, and / or the number of consecutive image frames without the second target object being identified reaches the third frame number; and maintain timing when, during the timing process, the duration of the second target object not being identified is less than the sixth duration, and / or the number of consecutive image frames without the second target object being identified is less than the third frame number. This avoids misidentification due to dropped image frames.

[0193] It is understandable that in practical applications, when identifying a second target object and determining the duration of its presence within a third distance range using acquired images, some frames may be lost due to factors such as external signal interference. This can easily lead to misjudgments of the duration the second target object has been within the third distance range, resulting in false alarms or missed alarms. By setting a sixth duration and a third frame number, misjudgments caused by image frame loss can be avoided.

[0194] Specifically, in some cases, the sixth duration can be a sixth set duration. The third frame number can be a third set frame number.

[0195] In this invention, the monitoring method may include:

[0196] If the second target object leaves the third distance range, determine whether there is still a second target object remaining in the fourth distance range of vehicle 100, where the fourth distance range is greater than or equal to the third distance range;

[0197] If it is determined that there is no second target within the fourth distance range and the duration reaches the seventh time, the alarm status of vehicle 100 is deactivated.

[0198] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4The control module 111 can be used to: determine whether a second target object remains within a fourth distance range of the vehicle 100 when the second target object leaves the third distance range, wherein the fourth distance range is greater than or equal to the third distance range; and deactivate the alarm status of the vehicle 100 if it is determined that no second target object exists within the fourth distance range and the determination is maintained for a duration of seven hours. This improves accuracy in determining that no second target object exists within the immediate vicinity of the vehicle 100.

[0199] It is understandable that when the second target has moved a certain distance away from vehicle 100, there is no longer any potential risk posed by the second target, and therefore the alarm does not need to be maintained. By setting the fourth distance range and the seventh duration, it can be determined that the second target is completely away from vehicle 100, thus facilitating the confirmation of when to deactivate the alarm.

[0200] Specifically, in some cases, the fourth distance range of vehicle 100 can be a circular area centered on vehicle 100 with a fourth set distance as the radius. The seventh duration can be a seventh set duration.

[0201] In this invention, the monitoring method may include:

[0202] If the second target first leaves the third distance range, start timing for the duration during which no second target exists within the fourth distance range;

[0203] During the timing process, if the duration of the second target object not being identified reaches the eighth duration, and / or the number of consecutive image frames without the second target object being identified reaches the fourth frame count, the timing shall be stopped.

[0204] If, during the timing process, the duration of the period during which the second target object is not continuously identified is less than the eighth duration, and / or the number of consecutive image frames in which the second target object is not identified is less than the fourth frame number, the timing is maintained.

[0205] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: start timing the duration for which no second target exists within a fourth distance range when the second target first leaves the third distance range; stop timing when the duration of the continuous absence of the second target reaches an eighth duration, and / or the number of consecutive image frames without the second target reaches a fourth frame count; and maintain timing when the duration of the continuous absence of the second target is less than an eighth duration, and / or the number of consecutive image frames without the second target is less than a fourth frame count. This avoids misidentification due to dropped image frames.

[0206] It is understandable that in practical applications, when identifying the second target object and determining the duration of its departure from vehicle 100 using acquired images, some image frames may be lost due to factors such as interference from external signals. This can easily lead to false alarms or missed alarms due to misjudgment of the duration of the second target object's departure from vehicle 100. By setting an eighth duration and a fourth frame number, false alarms caused by image frame loss can be avoided.

[0207] Specifically, in some cases, the eighth duration can be the eighth set duration. The fourth frame number can be the fourth set frame number.

[0208] In this invention, the monitoring method may include:

[0209] If the distance fluctuation of the second target relative to vehicle 100 is less than the preset fluctuation distance and the duration reaches the ninth time, the alarm status of vehicle 100 is deactivated.

[0210] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4 The control module 111 can be used to: deactivate the alarm status of vehicle 100 when the distance fluctuation amplitude of the second target object relative to vehicle 100 is less than a preset fluctuation distance and the duration reaches the ninth time. In this way, it can be determined that the second target object is stationary relative to vehicle 100 and does not pose a potential risk.

[0211] It is understandable that objects that move to the vicinity of vehicle 100 and then remain stationary can be considered to no longer pose a potential risk to vehicle 100, thus the alarm can be deactivated. In some scenarios, the second target object can be other vehicles parked within the vicinity of vehicle 100. By setting a preset fluctuation distance and a ninth duration, it is easy to determine whether the second target object remains stationary relative to vehicle 100.

[0212] Specifically, in some cases, the preset fluctuation distance can be the fifth preset distance. The ninth duration can be the ninth preset duration.

[0213] In this invention, the monitoring method may include:

[0214] If the alarm state of vehicle 100 is triggered, determine to record video of the scene where vehicle 100 is located.

[0215] The monitoring method of the present invention can be implemented using the vehicle 100 of the present invention. Specifically, please refer to... Figure 4The control module 111 can be used to: determine to record video of the scene where vehicle 100 is located when the alarm state of vehicle 100 is triggered. In this way, the specific process of vehicle 100 being collided or scraped can be recorded in a timely manner.

[0216] Specifically, please combine Figure 4 In some cases, the image sensor 112 can continuously acquire images. When the image determines that the vehicle 100 has a potential risk and triggers the alarm state of the vehicle 100, the image sensor 112 can be called to continue acquiring images for video recording, so that the specific process can be recorded in a timely manner when the vehicle 100 is damaged.

[0217] Please refer to Figure 5 A vehicle 100 according to the present invention may include a memory 201 and a processor 202. The memory 201 may store a computer program. When the processor 202 executes the computer program, it can implement the steps of the monitoring method of the present invention.

[0218] For example, when a computer program is executed by processor 202, the monitoring methods that can be implemented include:

[0219] 011: Acquire images of the scene where vehicle 100 is located;

[0220] 012: If it is determined that a first target object is staying within a first distance range of the vehicle 100 and the staying time reaches a first duration, the alarm state of the vehicle 100 is triggered. The first target object can be identified by image recognition and the first target object has human characteristics.

[0221] If the vehicle 100 determines that the first target object stays at a relatively close distance to the vehicle 100 for a long time and has human characteristics, it can be considered that someone has stayed around the vehicle 100 for a long time and may come into contact with the vehicle 100, posing a potential risk to the vehicle 100. This triggers an alarm state to facilitate further monitoring and reduce the number of missed alarms.

[0222] The present invention provides a computer-readable storage medium on which a computer program can be stored. When executed by a processor, the computer program can implement the steps of the monitoring method of the present invention.

[0223] For example, when a computer program is executed by a processor, the monitoring methods that can be implemented include:

[0224] 011: Acquire images of the scene where vehicle 100 is located;

[0225] 012: If it is determined that a first target object is staying within a first distance range of the vehicle 100 and the staying time reaches a first duration, the alarm state of the vehicle 100 is triggered. The first target object can be identified by image recognition and the first target object has human characteristics.

[0226] The aforementioned computer-readable storage medium, by determining that the first target object stays at a relatively close distance to the vehicle 100 for a relatively long time and has human characteristics, can be considered to indicate that someone has been staying around the vehicle 100 for a long time and may come into contact with the vehicle 100, posing a potential risk to the vehicle 100, thereby triggering an alarm state to facilitate further monitoring and reduce the number of missed alarms.

[0227] The computer-readable storage medium can be located in the control module 111 or in other terminals. The control module 111 can communicate with other terminals to obtain the corresponding program.

[0228] It is understood that computer-readable storage media can include: any entity or device capable of carrying a computer program, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc. Computer programs can include computer program code. Computer program code can be in the form of source code, object code, executable files, or certain intermediate forms, etc.

[0229] In some embodiments of the present invention, the control module 111 may be a microcontroller chip that integrates a processor, memory, communication module, etc. The processor may be a central processing unit (CPU), a graphics processing unit (GPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0230] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0231] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a system including a processing module or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0232] Although embodiments of the present invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to the embodiments of the present invention without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for monitoring vehicles, characterized in that, include: Acquire images of the scene where the vehicle is located; If it is determined that a first target object is staying within a first distance range of the vehicle and the staying time reaches a first duration, the alarm state of the vehicle is triggered. The first target object can be identified by image recognition and the first target object has human characteristics. The monitoring method includes: When the first target object first enters the first distance range of the vehicle, the duration of the first target object's stay is timed. During the timing process, if the duration during which the first target object is not continuously identified reaches a second duration, and / or the number of consecutive image frames in which the first target object is not identified reaches a first frame number, the timing is stopped. During the timing process, if the duration during which the first target object is not continuously identified is less than the second duration, and / or the number of consecutive image frames in which the first target object is not identified is less than the first number of frames, the timing is maintained.

2. The monitoring method according to claim 1, characterized in that, The monitoring method includes: If the first target object leaves the first distance range, determine whether the first target object is still within a second distance range of the vehicle, where the second distance range is greater than or equal to the first distance range; If it is determined that the first target is not present within the second distance range of the vehicle, and the duration of this condition reaches a third duration, the alarm status of the vehicle is deactivated.

3. The monitoring method according to claim 2, characterized in that, The monitoring method includes: When the first target object leaves the first distance range for the first time, start timing the duration during which the first target object does not remain within the second distance range; During the timing process, if the duration during which the first target object is not continuously identified reaches the fourth duration, and / or the number of consecutive image frames in which the first target object is not identified reaches the second frame number, the timing is stopped. During the timing process, if the duration during which the first target object is not continuously identified is less than the fourth duration, and / or the number of consecutive image frames in which the first target object is not identified is less than the second number of frames, the timing is maintained.

4. The monitoring method according to claim 1, characterized in that, The monitoring method includes: Obtain object information obtained by radar scanning of the scene where the vehicle is located; If a second target is determined to have entered the third distance range of the vehicle, and the second target has been within the third distance range for a duration of a fifth duration, the vehicle's alarm state is triggered. The second target can be identified through the object information, and the third distance range is smaller than the first distance range.

5. The monitoring method according to claim 4, characterized in that, The monitoring method includes: When the second target object first enters the third distance range, the duration of the second target object being within the third distance range is started. During the timing process, if the duration of the second target object not being identified reaches the sixth duration, and / or the number of consecutive image frames in which the second target object is not identified reaches the third frame number, the timing shall be stopped. During the timing process, if the duration during which the second target object is not continuously identified is less than the sixth duration, and / or the number of consecutive image frames in which the second target object is not identified is less than the third number of frames, the timing is maintained.

6. The monitoring method according to claim 4, characterized in that, The monitoring method includes: If the second target object leaves the third distance range, determine whether the second target object is still within the fourth distance range of the vehicle, where the fourth distance range is greater than or equal to the third distance range; If it is determined that there is no second target object within the fourth distance range and the duration reaches the seventh time, the alarm status of the vehicle is deactivated. And / or, the monitoring method includes: If the distance fluctuation between the second target and the vehicle is less than the preset fluctuation distance and the duration reaches the ninth hour, the alarm status of the vehicle is deactivated.

7. The monitoring method according to claim 6, characterized in that, The monitoring method includes: When the second target object first leaves the third distance range, the timer starts to count the duration during which the second target object is not present in the fourth distance range; During the timing process, if the duration of the second target object not being identified reaches the eighth duration, and / or the number of consecutive image frames in which the second target object is not identified reaches the fourth frame number, the timing shall be stopped. During the timing process, if the duration during which the second target object is not continuously identified is less than the eighth duration, and / or the number of consecutive image frames in which the second target object is not identified is less than the fourth number of frames, the timing is maintained.

8. The monitoring method according to any one of claims 1-7, characterized in that, The monitoring method includes: If the vehicle's alarm status is triggered, determine to record video of the scene where the vehicle is located.

9. A vehicle, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the monitoring method according to any one of claims 1-8.

Images