A sentry mode alert state triggering method, device and vehicle

Abstract

The embodiment of the application provides a kind of sentinel mode vigilance state triggering method, device and vehicle, it is related to vehicle monitoring technical field, including in the monitoring state of sentry system, at least including the distance and the multiple behavior data of hand and the current environmental factor information of surrounding personnel are acquired continuously, wherein the behavior state detection of personnel is the main computing factor of risk event occurrence probability, and environmental factors are associated with the influence of human behavior and environment, in order to reduce the prediction result of the risk of damage to the car that may occur misjudgment, personnel factors are corrected by environmental factors, so as to be able to be based on more accurate risk probability prediction, decision whether to trigger sentry system to enter into vigilance state.Compared with the existing sentry state triggering form, the application can more reliably and accurately enter the vigilance state, while ensuring to reduce unnecessary resource consumption, so that the vigilance mechanism of sentry system plays a practical effect.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of vehicle monitoring, in particular to a method and device for triggering a vigilant state of a sentry mode and a vehicle. BACKGROUND

[0002] There are generally three states of a sentry mode of a vehicle: a monitoring state, a vigilant state and an alarm state. The monitoring state means that the sentry system continuously works to monitor the environment around the vehicle, but does not trigger an alarm. The vigilant state means that when the sentry system detects, through sensors such as ring-view cameras and ultrasonic radars, that the vehicle is at risk of being damaged or harmed, it outputs a first-level alarm, for example, timely illuminates the center control screen to alert the personnel intending to damage the vehicle that the vehicle is in a state of continuously recording monitoring videos. The alarm state means that when the sentry system detects that the vehicle has been damaged, it immediately issues a second-level alarm, such as a sound-light alarm or a networked alarm.

[0003] Since the sentry mode needs to continuously work after the vehicle is powered off, a high power consumption state is a key problem in the industry. It is verified in practice that the power consumption is relatively large when in the aforementioned vigilant state and alarm state, and in particular, the triggering judgment of the vigilant state is relatively unclear and complex compared to the triggering condition of the alarm state, which often leads to unnecessary resource waste caused by false triggering into the vigilant state. SUMMARY

[0004] Embodiments of the present application provide a method and device for triggering a vigilant state of a sentry mode and a vehicle, which are related to the technical field of vehicle monitoring. The method can combine various behavior states of personnel outside the vehicle and current environmental factors, comprehensively calculate the risk probability of causing damage to the vehicle, and decide whether to trigger the sentry vigilant state to remind high-risk personnel, so as to more reliably and accurately enter the vigilant state compared to the existing sentry state triggering form, while reducing unnecessary resource consumption, and make the vigilant mechanism of the sentry system effective.

[0005] In a first aspect, embodiments of the present application provide a method for triggering a vigilant state of a sentry mode, the method comprising:

[0006] After entering the sentry mode, detecting various behavior states of personnel around the vehicle and a current environmental state, the behavior states at least including: a distance between the personnel and the vehicle, and a hand state of the personnel;

[0007] Using the detected behavior state data, calculating a probability value of a preset risk event corresponding to each of the behavior states;

[0008] Based on a preset first weight corresponding to each of the behavior states, fusing the probability values of all the behavior states to obtain a total probability value;

[0009] adjust the total probability value in combination with the detected environment state, and decide whether to trigger the entry into the alert state according to the adjusted total probability value.

[0010] The technical effects of the scheme can be referred to as follows. In the monitoring state of the sentinel system, the surrounding personnel are continuously acquired to include at least a plurality of behavior data of distance and hand and current environment factor information. The behavior state detection of the personnel is a main calculation factor of the risk event occurrence probability, and the environment factor is correlated considering the influence of the human behavior and the environment. In order to reduce the prediction result of the risk of damage to the vehicle caused by misjudgment, the personnel factor is corrected by the environment factor, so as to be able to predict the risk probability more accurately, and decide whether to trigger the sentinel system to enter the alert state. Compared with the existing sentinel state triggering form, the application can more reliably and accurately enter the alert state, while ensuring to reduce unnecessary resource consumption, so that the alert mechanism of the sentinel system plays a practical effect.

[0011] In at least one possible implementation, the triggering method further includes: fusing a plurality of the environment states based on preset second weights corresponding to each of the environment states to obtain an adjustment parameter for adjusting the total probability value.

[0012] In the embodiment, for a plurality of environment factors that may affect the behavior, a weight fusion method is used to integrate different environment states into a unified correction parameter of the total probability value, so as to facilitate the adjustment of the total probability value.

[0013] In at least one possible implementation, the environment state includes at least one of the following: a parking place, a parking area monitoring site, an illumination intensity of the parking area, a date and time, and historical data of non-safety events in the parking area.

[0014] In the embodiment, the environment factors are expanded from multiple dimensions such as space, time, illumination, external facilities, and risk event history, and external condition factors that may affect the behavior of the personnel are proposed, so as to improve the prediction accuracy of the risk degree of the personnel behavior.

[0015] In at least one possible implementation, the triggering method further includes: outputting a preset warning audio for reminding the personnel outside the vehicle after the sentinel mode decides to enter the alert state.

[0016] In the embodiment, the preset audio is emphasized as the execution result of the alert state of the sentinel, which plays a role in energy saving on the one hand and a better warning and reminding role on the other hand.

[0017] In at least one possible implementation, the warning audio includes at least one of the following: an audio simulating engine starting, and a warning voice.

[0018] In this embodiment, with a specific warning audio type, it is ensured that in a non-alarm situation, the purpose of effectively avoiding risks is achieved.

[0019] In at least one possible implementation, the behavior state further includes at least one of: a moving direction of the person, a moving speed of the person, a face orientation of the person, a line-of-sight orientation of the person, and a staying time of the person.

[0020] In addition to the two main behavior states mentioned above, this embodiment also provides detection of other behavior states that may be associated with risk behaviors, increases the reference dimension of risk probability estimation, and assists in improving the reliability of subsequent triggering into the alert state and reducing the probability of false triggering.

[0021] In a second aspect, the embodiments of the present application provide a sentinel mode alert state triggering device, which comprises:

[0022] a behavior environment detection module, configured to detect a plurality of behavior states of persons around the vehicle and a current environment state after entering the sentinel mode, wherein the behavior states include at least a distance between the persons and the vehicle and a hand state of the persons;

[0023] a risk event probability calculation module, configured to calculate a probability value of a preset risk event corresponding to each of the behavior states by using the detected behavior state data;

[0024] a probability fusion module, configured to fuse the probability values of all the behavior states to obtain a total probability value based on a preset first weight corresponding to each of the behavior states;

[0025] a state triggering decision module, configured to adjust the total probability value in combination with the detected environment state, and decide whether to trigger into the alert state according to the adjusted total probability value.

[0026] In a third aspect, the embodiments of the present application provide an electronic device, which comprises one or more processors, a memory, and one or more computer programs, wherein the memory can be a non-volatile storage medium, the one or more computer programs are stored in the memory, and the one or more computer programs comprise instructions, which, when executed by the device, cause the electronic device to perform the method in the first aspect or any possible implementation manner of the first aspect.

[0027] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores a computer program, and when the computer program is run on a computer, the computer is caused to perform the method in the first aspect or any possible implementation manner of the first aspect.

[0028] In a fifth aspect, an embodiment of the present application provides a vehicle, the vehicle comprising the electronic device in the third aspect, and the computer readable storage medium in the fourth aspect.

[0029] It should be understood that the second to fifth aspects of the embodiments of the present application are consistent with the technical solutions of the first aspect of the embodiments of the present application, and the beneficial effects obtained by each aspect and the corresponding feasible implementation manners are similar, which will not be described again.

DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present specification, and other drawings can also be obtained by those skilled in the art without creative labor.

[0031] Figure 1 A flowchart of a sentinel mode alert state triggering method provided by an embodiment of the present application;

[0032] Figure 2 A structural diagram of a sentinel mode alert state triggering device provided by an embodiment of the present application.

DETAILED DESCRIPTION

[0033] In order to better understand the technical solutions of the present specification, the embodiments of the present application will be described in detail below with reference to the drawings.

[0034] It should be clear that the described embodiments are only some embodiments of the present specification, not all embodiments. Based on the embodiments in the present specification, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present specification.

[0035] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present specification. The singular forms "a", "an" and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0036] At present, the sentinel mode alert state triggering condition is relatively single, which often causes the vehicle to enter the alert state and light up the screen in the vehicle, which not only wastes unnecessary electric energy, but also cannot reliably and accurately play the role of warning risks. Especially for the working condition scene with people coming and going around the vehicle, the mis-triggering situation occurs from time to time.

[0037] In view of this, the embodiment of the present application provides a sentinel mode alert state triggering method, and the main design idea is that in the monitoring state of the sentinel system, the surrounding personnel are continuously acquired at least including distance and hand behavior data and current environmental factor information, wherein the behavior state detection of the personnel is the main calculation factor of the risk event occurrence probability, and the environmental factor is related to the influence of the behavior of the personnel and the environment, and in order to reduce the prediction result of the risk of damage to the vehicle caused by misjudgment, the personnel factor is corrected by the environmental factor, so as to be able to predict the risk probability more accurately, and to decide whether to trigger the sentinel system to enter the alert state. Compared with the existing sentinel state triggering form, the present application can more reliably and accurately enter the alert state, while reducing unnecessary resource consumption, and making the alert mechanism of the sentinel system have a practical effect.

[0038] The technical solutions protected by the embodiment of the present application will be described in detail below with reference to the drawings.

[0039] Please refer to Figure 1 A flowchart of a sentinel mode alert state triggering method provided by the embodiment of the present application is shown. The flowchart of the method is described as follows.

[0040] Step 101: After entering the sentinel mode, the behavior state of the personnel around the vehicle and the current environmental state are detected, and the behavior state at least includes the distance between the personnel and the vehicle and the hand state of the personnel.

[0041] It is considered that in a real situation, when the external personnel damage the vehicle, there are mainly two key factors, one is that the personnel (pedestrians or personnel driving other vehicles, etc.) around the vehicle and the vehicle need to be within a certain distance range; the second is that the behavior of damaging the vehicle is strongly related to the hand state of the personnel (for example, whether the hand holds an article, the hand posture, etc.). Therefore, in the embodiment, the two key factors are mainly proposed to be investigated, and in actual operation, the investigation method can not be limited to using the existing machine learning algorithm model, and the image information obtained in the monitoring state of the sentinel mode. Those skilled in the art can understand that the technical implementation means of detecting the behavior state of the personnel around the vehicle is not the focus of the present application, and many mature algorithm models can be used for reference, and the present embodiment will not be repeated and limited.

[0042] It is worth noting that in addition to predicting whether there is a risk in the comprehensive behavior state of the person, the embodiment also mentions the consideration of environmental factors. Generally speaking, the object of the sentry mode is "people", but the analysis of the present application shows that environmental factors also play an important role in the behavior risk of "people". Therefore, it is proposed to add the environmental state of the sensing object for the detection of the sentry mode. Specifically, in some preferred embodiments, the environmental state at least includes one of the following: parking place (such as underground or indoor parking lot, outdoor parking lot, etc.), parking area monitoring site (the number, position and coverage of external monitoring of the current parking area can be obtained through image acquisition, data networking or navigation map, etc.), light intensity of the parking area (such as judging whether there is light or in the daytime), date and time (which can also involve daytime judgment, and can also include factors such as holidays and seasons); of course, in other embodiments, the environmental state can also include: non-safety event historical data related to vehicle damage in the current parking area obtained through networked data, for example, from the data information published by the traffic management platform, police platform, etc. The number of events, incidence, trend, etc. of human-caused vehicle property damage in this area within a certain period of time in the past.

[0043] It can be added here that in addition to the above behavior state, in other embodiments of the present application, further consideration is also given to behaviors that may cause damage or destruction of vehicles in real situations, which can include at least one of the following: personnel moving direction, personnel moving speed, personnel face orientation, personnel line of sight orientation, personnel stay time, etc.

[0044] Step 102: using the detected behavior state data, calculating the probability value of the preset risk event corresponding to each of the behavior states.

[0045] It can be pointed out here that since the present application focuses on the possibility of the off-vehicle person implementing damage, destruction and other behaviors to the vehicle, the risk events corresponding to each behavior state can be set in advance based on different behavior states in combination with the established behavior threshold value, etc. For example, the risk event corresponding to the distance between the person and the vehicle is that the person is close to the vehicle (the distance can be determined according to the distance threshold value), for example, the risk event corresponding to the hand state of the person can include holding an object (which can be determined according to the target recognition algorithm) or dynamically approaching the vehicle (which can be determined according to the motion detection algorithm), for example, the risk event corresponding to the line of sight of the person can be set to (which can be determined according to the number of times and the time length of the line of sight staring at the vehicle within a predetermined period of time), and so on. For each behavior state mentioned or not mentioned above, the corresponding risk event can be set. Then, the algorithm in the field of probability statistics such as Bayesian network can be used to calculate the probability value of each risk event leading to the alert state and triggering the first risk alarm. Some examples of incremental probability calculation schemes are introduced in the following, which will not be repeated here.

[0046] Step 103: Based on the preset first weight corresponding to each behavior state, the probability values of all behavior states are fused to obtain a total probability value.

[0047] The first weight mentioned in this link can set the specific weight value of each behavior state according to the key degree of the event factor triggering the alert state as mentioned above. The way to fuse the probability values can be but not limited to realized by weighted summation.

[0048] Step 104: Adjust the total probability value in combination with the detected environment state, and decide whether to enter the alert state according to the adjusted total probability value.

[0049] It can be unfolded that, since the embodiment not only detects the behavior of the person outside the vehicle in the sentinel mode, but also detects the environmental data affecting the behavior of the person, after obtaining the total score of the probability based on the multi-dimensional behavior state, it is proposed to use the current environmental state detection result to correct the total score, that is, not only the behavior of the person is relied on as the basis for decision-making to enter the sentinel alert state, but also environmental factors such as the aforementioned parking place, light intensity, external monitoring facilities, date and time, and even high-risk event historical data are fused, which is very helpful to improve the accuracy of predicting the risk possibility. Based on this concept, it can also be supplemented that if more than one of the environmental states is detected, a second weight corresponding to each of the environmental states can be preset based on the detected overall environmental state to obtain the adjusted parameter after the aggregation. That is, the influence degree of different environmental factors can be fused to adjust the total probability value; and the specific adjustment method can be embodied in the form of a coefficient or can be embodied in the form of positive and negative compensation of the probability value, which is not limited by the embodiment. In addition, it can also be understood that the basis for decision-making in this step is the comparison relationship between the total probability value adjusted by the environmental factors and the established probability threshold, which is also not limited.

[0050] Here, in combination with the foregoing embodiment and the Bayesian network, the following example is provided for reference. It should be understood that the following example only takes the four behavior states of the person mentioned in the foregoing as a schematic description of the probability calculation process (the adjustment mechanism of other behavior factors and environmental states is not involved):

[0051] It is assumed that A1 event is that the person is close to the vehicle, A2 event is that the person moves towards the vehicle, A3 event is that the person moves fast, and A4 event is that the person's face is continuously directed at the vehicle; B represents the occurrence of a first-level risk, that is, entering the alert state of the sentinel mode.

[0052] A1 event probability: P(A1) = 0, if L > 2 meters; P(A1) = 1, if L < 0.2 meters; P(A1) = (L-0.2) / (2-0.2), if 2 meters ≥ L ≥ 0.2 meters, where L is the distance between the pedestrian and the vehicle.

[0053] A2 event probability: P(A2) = 0, if the direction of the pedestrian's movement is not towards the vehicle in the past 5 seconds; P(A2) = 1, if the direction of the pedestrian's movement is always towards the vehicle in the past 5 seconds; P(A2) = t / 5, if the direction of the pedestrian's movement is towards the vehicle for t seconds in the past 5 seconds.

[0054] A3 event probability: P(A3) = 0, in the past 5 seconds, the average moving speed of the pedestrian v < 0.1 m / s; P(A3) = 1, in the past 5 seconds, the average moving speed of the pedestrian v > 1.5 m / s; P(A3) = v / (1.5_0.1). In the past 5 seconds, the average moving speed of the pedestrian v is between 0.1 and 1.5.

[0055] A4 event probability: P(A4) = 0, in the past 5 seconds, the face of the pedestrian is not facing the vehicle; P(A4) = 1, in the past 5 seconds, the face of the pedestrian is always facing the vehicle; P(A4) = t / 5. In the past 5 seconds, the face of the pedestrian is facing the vehicle for t seconds.

[0056] It should be additionally pointed out that the above is only an example of probability calculation, and other incremental probability schemes can be adopted.

[0057] Then, according to the pre-set directed edge weight value, that is, according to the importance of the factors of the vehicle being damaged and destroyed by external personnel to the first risk level, the weights of A1→B, A2→B, A3→B and A4→B are designed respectively. Those skilled in the art can understand that the weight value can be a calibrated value and can be adjusted as needed. In the above example, α, β, γ, δ are used to represent respectively, then the probability of combining multiple behavior states is calculated P(B): P(B) = α*P(A1) + β*P(A2) + Y*P(A3) + δ*P(A4).

[0058] Then, according to the threshold value, if P(B) > θ, it is determined that a first risk event occurs (entering an alert state). Wherein, θ can also be a calibrated value adjusted as needed.

[0059] In the foregoing, after the sentinel mode decision enters the alert state, at least a preset warning audio is output to remind the possibility of high-risk behavior of the person outside the vehicle. The present embodiment proposes to use the preset warning audio as the execution content of the alert state, mainly considering two aspects: first, compared with the conventional screen lighting reminder, audio output consumes less energy; second, for the scene concerned in the present application, the effect of sound reminder is better than that of screen lighting, for example, in other better embodiments, the warning audio at least includes one of the following: audio simulating engine (engine or motor) start, voice output according to a preset warning template (for reference, "the vehicle is in video monitoring state", "the vehicle has been connected to the police platform", etc.). Compared with screen lighting, such audio output reminder information can better achieve the warning effect of making the person outside the vehicle stop the risk behavior (such as the person who is too close to the vehicle hearing the above audio will have a greater probability of moving away from the vehicle), and even deter some illegal behavior (the person holding a weapon approaching the vehicle from a distance and focusing on the vehicle door will implement the destruction behavior).

[0060] It can also be added that the aforementioned warning audio, which is different from the secondary alarm sound output by the alert state of the sentry mode, is an audio for reminding the warning, because the determination of the alert stage is derived from the prediction that is not explicit, and thus needs to be mainly reminded of the warning.

[0061] Referring to Figure 2 Based on the same inventive concept, the embodiment of the present application also provides a sentry mode alert state triggering device, which comprises:

[0062] The behavior environment detection module 201 is configured to detect a plurality of behavior states of the people around the vehicle and a current environment state after entering the sentry mode, wherein the behavior states at least include the distance between the people and the vehicle and the hand state of the people.

[0063] The risk event probability calculation module 202 is configured to calculate the probability value of a preset risk event corresponding to each of the behavior states by using the detected behavior state data.

[0064] The probability fusion module 203 is configured to fuse the probability values of all the behavior states to obtain a total probability value based on a preset first weight corresponding to each of the behavior states.

[0065] The state triggering decision module 204 is configured to adjust the total probability value in combination with the detected environment state, and decide whether to trigger the entry into the alert state according to the adjusted total probability value.

[0066] Based on the same inventive concept, the embodiment of the present application also provides an electronic device, which comprises at least one processor, and the processor is configured to execute a computer program stored in a memory to implement the flowchart steps of the sentry mode alert state triggering method provided by the embodiment of the present application.

[0067] Optionally, the processor can be a central processor, a specific ASIC, and can be one or more integrated circuits for controlling program execution.

[0068] Optionally, the electronic device can further comprise a memory connected to the at least one processor, and the memory can include a ROM, a RAM and a disk memory. The memory is used to store the data required by the processor during execution, that is, the instructions executable by the at least one processor are stored, and the at least one processor executes the instructions stored in the memory to execute the method mentioned in each of the above embodiments. Wherein, the number of memories is one or more.

[0069] The embodiment of the present application further provides a computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions run on a computer, the computer executes the method mentioned in the above embodiments.

[0070] Based on the same inventive concept, the embodiment of the present application further provides a vehicle comprising at least the electronic device and / or the computer readable storage medium. The selection of the vehicle can be determined by the correct understanding and reasonable implementation of the foregoing scheme by the person skilled in the art.

[0071] The above merely describes preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of alert state triggering in a sentinel mode, characterized by, The method comprises: After entering the sentinel mode, detecting a plurality of behavior states of people around the vehicle and a current environment state, the behavior states at least including: a distance between the people and the vehicle, and a hand state of the people; Using the detected behavior state data, calculating a probability value of a preset risk event corresponding to each of the behavior states; Based on a preset first weight corresponding to each of the behavior states, fusing the probability values of all the behavior states to obtain a total probability value; Adjusting the total probability value in combination with the detected environment state, and deciding whether to trigger the alert state according to the adjusted total probability value.

2. The sentinel mode alert state triggering method of claim 1, wherein, The triggering method further comprises: based on a preset second weight corresponding to each of the environment states, fusing a plurality of the environment states to obtain an adjustment parameter for adjusting the total probability value. 3.The sentinel mode alert state triggering method of claim 1, wherein, The environment state at least includes one of the following: a parking location, a parking area monitoring site, an illumination intensity of the parking area, a date and time, and non-safety event historical data of the parking area.

4. The sentinel mode alert state triggering method of claim 1, wherein, The triggering method further comprises: after deciding to enter the alert state in the sentinel mode, outputting a preset warning audio for reminding people outside the vehicle.

5. The sentinel mode alert state triggering method of claim 4, wherein, The warning audio at least includes one of the following: an audio simulating engine starting, and a warning voice.

6. The sentinel mode alert state triggering method according to any one of claims 1 to 5, characterized in that, The behavior state at least further includes one of the following: a moving direction of the people, a moving speed of the people, a face orientation of the people, a line-of-sight orientation of the people, and a staying time of the people.

7. A vigilant state triggering apparatus in a sentinel mode, characterized by, The device comprises: A behavior environment detection module for detecting a plurality of behavior states of people around the vehicle and a current environment state after entering the sentinel mode, the behavior states at least including: a distance between the people and the vehicle, and a hand state of the people; A risk event probability calculation module for calculating a probability value of a preset risk event corresponding to each of the behavior states using the detected behavior state data; A probability fusion module for fusing the probability values of all the behavior states based on a preset first weight corresponding to each of the behavior states to obtain a total probability value; A state triggering decision module for adjusting the total probability value in combination with the detected environment state, and deciding whether to trigger the alert state according to the adjusted total probability value.

8. An electronic device, comprising: Comprise: One or more processors, memories, and one or more computer programs, wherein the one or more computer programs are stored in the memories, the one or more computer programs comprising instructions that, when executed by the electronic device, cause the electronic device to perform the alert state triggering method of the sentinel mode according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the alert state triggering method of the sentinel mode according to any one of claims 1-6.

10. A vehicle characterized by comprising: The vehicle is configured with the electronic device according to claim 8 or the computer readable storage medium according to claim 9.

Images