Method and device for controlling start and stop of sentry mode and electronic equipment

Abstract

The embodiment of the invention relates to the technical field of vehicles, and provides a sentry mode start-stop control method and device and electronic equipment. The method comprises the following steps: when a sentry mode is started, acquiring an initial image corresponding to each image channel; obtaining environment information of the vehicle according to the initial image corresponding to each image channel; and closing an image channel corresponding to at least one target direction in response to detecting that an obstacle exists outside the vehicle in the environment information and the position relationship between the obstacle and the vehicle in the at least one target direction meets a preset position condition, so as to solve the problem of high power consumption in a sentry mode of the vehicle.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of vehicles, and particularly relates to a control method and device for starting and stopping a sentry mode and an electronic device. BACKGROUND

[0002] At present, the sentry mode of a vehicle has become an important function in an intelligent cockpit and a vehicle safety system, which effectively prevents potential damage or theft by continuously monitoring the surrounding environment of the vehicle in a parking state. However, in the actual application process, the sentry mode of the vehicle faces a significant power consumption problem, which restricts its large-scale promotion.

[0003] After the sentry mode is started, all sensing channels are continuously running and are analyzed, but in some special scenarios, the data collected by part of the image channels cannot be effectively used for environment perception, causing unnecessary waste of resources. SUMMARY

[0004] The sentry mode of the vehicle, in a parking state, monitors the vehicle and the surrounding environment through various sensors and cameras of the vehicle body, and obtains the safety situation of the vehicle in real time. All image channels around the vehicle are continuously running, which has extremely high power consumption. However, detection in some scenarios is meaningless, for example, there is no risk or extremely low risk on the side of the vehicle close to the fence, and continuous detection is not needed.

[0005] To solve the above technical problems, the present disclosure provides a control method and device for starting and stopping a sentry mode and an electronic device to solve the problem of high power consumption of the sentry mode of the vehicle.

[0006] A first aspect of the present disclosure provides a control method for starting and stopping a sentry mode, comprising:

[0007] When the sentry mode is started, initial images corresponding to each image channel are obtained;

[0008] According to the initial images corresponding to each image channel, environment information of the vehicle is obtained;

[0009] In response to detecting that an obstacle exists outside the vehicle in the environment information, and a positional relationship between the obstacle and the vehicle in at least one target direction satisfies a preset position condition, an image channel corresponding to the at least one target direction is closed.

[0010] A second aspect of the present disclosure provides a control device for starting and stopping a sentry mode, comprising:

[0011] An image acquisition component comprising a plurality of cameras outside the vehicle, the cameras being configured to, when the sentry mode is started, acquire initial images outside the vehicle through the cameras corresponding to the image channels;

[0012] one or more processors configured to execute instructions stored in the memory to obtain the environmental information of the vehicle according to the initial image corresponding to each image channel; in response to detecting that there is an obstacle outside the vehicle in the environmental information, and that the positional relationship between the obstacle and the vehicle in at least one target direction meets a preset positional condition, closing the image channel corresponding to the at least one target direction.

[0013] In a third aspect of the present disclosure, an electronic device is provided, which includes one or more processors, and a memory; the memory stores computer instructions; the computer instructions, when executed by the processor, cause the processor to perform the control method for starting and stopping the sentry mode according to the first aspect.

[0014] In a fourth aspect of the present disclosure, a computer readable storage medium is provided, which stores computer program instructions; the computer program instructions, when executed by a processor, cause the processor to perform the control method for starting and stopping the sentry mode according to the first aspect.

[0015] Based on the control method for starting and stopping the sentry mode provided in the present disclosure, when the sentry mode is started, the initial image corresponding to each image channel is obtained and the environmental information of the vehicle is analyzed to identify the obstacle existing outside the vehicle. When the positional relationship between the obstacle and the vehicle in a specific target direction meets a preset positional condition, the image channel corresponding to the target direction is closed. In this way, the continuous image acquisition and processing in the direction without risk or with extremely low risk are effectively avoided, the power consumption is significantly reduced, the monitoring time of the vehicle in the parking state is prolonged, and the practicability and user experience of the sentry mode are improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the figures. The drawings provided in the present disclosure are used to provide further understanding of the embodiments of the present disclosure, and constitute a part of the specification, and are used to explain the present disclosure together with the embodiments of the present disclosure, but do not constitute a limitation to the present disclosure. In the drawings, the same reference numerals represent the same or similar components throughout the figures.

[0017] Figure 1 is a structural schematic block diagram of a control system for starting and stopping the sentry mode provided by an exemplary embodiment of the present disclosure;

[0018] Figure 2 is a schematic diagram of an acquisition area provided by an exemplary embodiment of the present disclosure;

[0019] Figure 3 is a flowchart of a control method for starting and stopping the sentry mode provided by an exemplary embodiment of the present disclosure;

[0020] Figure 4Ais a schematic diagram of a position relationship between an obstacle and a vehicle provided by an example embodiment of the present disclosure;

[0021] Figure 4B is a schematic diagram of a position relationship between an obstacle and a vehicle provided by another example embodiment of the present disclosure;

[0022] Figure 5 is a flowchart of a control method for starting and stopping a sentinel mode provided by another example embodiment of the present disclosure;

[0023] Figure 6 is a flowchart of a control method for starting and stopping a sentinel mode provided by another example embodiment of the present disclosure;

[0024] Figure 7 is a flowchart of a determination method for image quality provided by an example embodiment of the present disclosure;

[0025] Figure 8 is a flowchart of an acquisition method for an image quality category provided by an example embodiment of the present disclosure;

[0026] Figure 9 is a schematic diagram of a position relationship between an obstacle and a vehicle provided by another example embodiment of the present disclosure;

[0027] Figure 10 is a schematic diagram of a length of an obstacle provided by an example embodiment of the present disclosure;

[0028] Figure 11 is a flowchart of a determination method for a relative position between an obstacle and a vehicle provided by an example embodiment of the present disclosure;

[0029] Figure 12 is a structural schematic diagram of a control device for starting and stopping a sentinel mode provided by an example embodiment of the present disclosure;

[0030] Figure 13 is a structural schematic diagram of an electronic device provided by an example embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] Hereinafter, example embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the example embodiments described herein.

[0032] It should be noted that: unless otherwise specified, the relative arrangement, numerical expression and numerical value of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0033] SUMMARY

[0034] Sentry mode, when a vehicle is stationary, continuously monitors its surroundings to effectively prevent potential vandalism or theft. This typically requires the vehicle to simultaneously utilize multiple cameras—front, rear, and side—for uninterrupted recording. The sensors, image processors, and related electronic systems of these cameras all require continuous power. To analyze the camera footage in real time and detect potential threats, such as approaching people or unusual movement, the vehicle's high-performance computing chip needs to remain operational. Therefore, while Sentry mode provides comprehensive vehicle security even when the vehicle is stationary, the continuous operation of these components results in extremely high power consumption.

[0035] Once Sentinel mode is activated, all sensor channels continue to operate and analyze data. However, in certain special scenarios, data from some image channels cannot be effectively used for environmental perception, resulting in energy waste. For example, when a vehicle is parked with one side against a fixed obstacle, that side can be considered safe. Therefore, continuous detection has limited significance.

[0036] This disclosure proposes a novel solution to address the high power consumption issue of the aforementioned sentry mode. Specifically, this disclosure provides a control method, apparatus, and electronic device for starting and stopping sentry mode. When sentry mode is activated, initial images corresponding to each image channel are acquired. Based on the acquired initial images corresponding to each image channel, environmental information of the vehicle is obtained. If the environmental information includes the presence of an obstacle outside the vehicle, and the relationship between the obstacle's target direction and the vehicle's position information satisfies a preset position condition, it can be considered that the vehicle is safe in the target direction of the obstacle. The overall power consumption in sentry mode can be reduced by shutting down the image channel in that target direction.

[0037] Exemplary System

[0038] Figure 1 This is a schematic diagram of the structure of a sentinel mode start-stop control system provided in an exemplary embodiment of this disclosure.

[0039] like Figure 1 As shown, in one embodiment, the control system for starting and stopping the sentry mode may include a data acquisition system 10 and a decision system 20. The data acquisition system 10 and the decision system 20 are communicatively connected, so that the data acquisition system 10 and the decision system 20 can interact with each other.

[0040] In one implementation, the acquisition system 10 is used to acquire environmental data around the vehicle, including video and / or images. The acquisition system 10 may include at least one camera, wherein the camera may include at least one of a monocular camera, a binocular camera, a tri-lens camera, a wide-angle camera, and a fisheye camera; the type of camera is not limited in this disclosure embodiment. Thus, the acquisition system 10 can acquire video or images of the area around the vehicle.

[0041] In one example, the images acquired by the acquisition system 10 may include objects around the vehicle, such as walls, fences, greenery, pedestrians, animals, and vehicles.

[0042] The data acquisition system 10 may include at least one camera, which may be mounted at different locations on the vehicle to collect environmental data from different directions. For example, the camera may be mounted at the front of the vehicle to collect environmental data in front of the vehicle. The camera may be mounted at the rear of the vehicle to collect environmental data behind the vehicle. The camera may be mounted on the side of the vehicle to collect environmental data to the side of the vehicle.

[0043] The camera's capture area refers to the maximum shooting range of the camera. The camera's capture area is determined by the camera's shooting parameters, such as the field of view (FOV) and shooting distance.

[0044] In one example, if the camera's field of view and shooting distance are small, the camera's capture area is small. If the camera's field of view and shooting distance are large, the camera's capture area is large.

[0045] Correspondingly, the total acquisition area of ​​the acquisition system 10 is the sum of the acquisition areas of each camera.

[0046] In one implementation, the total acquisition area of ​​the acquisition system 10 can be a ring-shaped area centered on the vehicle. This ensures more comprehensive acquisition of environmental data around the vehicle.

[0047] Figure 2 This is a schematic diagram of a collection area provided for an exemplary embodiment of the present disclosure.

[0048] like Figure 2 As shown in this embodiment, cameras are installed at the front, rear, left, and right sides of the vehicle. The image acquisition range of each camera includes fan-shaped areas 11-14, and the total acquisition area is obtained by combining fan-shaped areas 11-14. The images captured from each direction can be stitched together to obtain complete environmental information around the vehicle.

[0049] In an implementation manner, the decision system 20 can include one or more processors 201. The processor 201 can include a general-purpose processor such as a central processing unit (CPU), a graphics processing unit (GPU), etc., and can also include an acceleration computing unit designed for deep learning tasks, autonomous driving tasks, sentinel mode tasks, etc., such as a neural processing unit (NPU), etc.

[0050] In an implementation manner, the decision system 20 can further include one or more memories 202. The memory 202 can store program instructions executable by the processor 201. The processor 201 can load and execute the program instructions in the memory 202 to implement the functions of the decision system 20.

[0051] In addition, the memory 202 can also cache or store intermediate data or result data generated by the processor 201 during operation, and store system files, application files, data files, etc. For example, the memory 202 can store the environmental data collected by the collection system 10, etc.

[0052] For example, the memory 202 can include a volatile memory such as a dynamic random access memory (DRAM), a static random access memory (SRAM), etc., and can also include a non-volatile memory such as a read-only memory (ROM), a flash memory, etc.

[0053] In an implementation manner, the decision system 20 can interact with the collection system 10 to obtain image data collected by the collection system 10, identify an obstacle existing around the vehicle based on the image data, and control to close an image channel corresponding to a target direction in a case where a positional relationship between the obstacle and the vehicle in the target direction satisfies a preset position condition, so as to effectively reduce the overall power consumption of the sentinel mode.

[0054] The scheme related in the present disclosure is not limited to the above-mentioned embodiments.

[0055] Exemplary method

[0056] Figure 3is a flowchart of a control method of a sentinel mode start-stop provided by an example embodiment of the present disclosure. The embodiment can be applied to an electronic device, such as Figure 3 As shown in the figure, the method comprises the following steps:

[0057] Step S101, when the sentinel mode is started, initial images corresponding to each image channel are acquired.

[0058] The channel refers to a logical set of hardware acquisition components, data transmission links and software processing resources configured independently to achieve specific environmental perception functions in the vehicle sentinel mode. The image channel is a specific implementation form of the channel in the embodiment of the present disclosure, including a channel with optical image acquisition and processing as the core function. It can be composed of a dedicated camera, an independent image data transmission path, a dedicated image data buffer area, and independent image processing processes and computing resources to complete image signal acquisition, transmission and processing. That is, each image channel is a set of hardware and / or software resources for processing images. For example, multiple image channels can be set in the sentinel mode, and each image channel corresponds to a direction, such as an image channel corresponding to the front of the vehicle, an image channel corresponding to the rear of the vehicle, an image channel corresponding to the left side of the vehicle, and an image channel corresponding to the right side of the vehicle. The initial image is an image acquired by the camera corresponding to the image channel. For example, the camera can be a fisheye camera.

[0059] The technical solution provided in this step acquires initial images corresponding to each image channel when the sentinel mode of the vehicle is started, providing complete and synchronous data sources for subsequent analysis of environmental information, and avoiding misjudgment caused by incomplete images.

[0060] Step S102, obtaining environmental information of the vehicle according to the initial images corresponding to each image channel.

[0061] The environmental information is information of the external surrounding environment of the vehicle, which can be used to determine whether there is an obstacle outside the vehicle, the position of the obstacle, the size of the obstacle, the type of the obstacle, and the like.

[0062] The technical solution provided in this step obtains partial environmental information of each direction outside the vehicle according to the initial images corresponding to each image channel, and obtains complete environmental information around the vehicle by combining the partial environmental information of each direction, thereby providing a reliable basis for subsequent closing decisions of the image channel.

[0063] Step S103, in response to detecting that there is an obstacle outside the vehicle in the environmental information, and that the position relationship between the obstacle and the vehicle in at least one target direction satisfies a preset position condition, closing the image channel corresponding to the at least one target direction.

[0064] In the embodiments of the present disclosure, the obstacle can be a fixed object outside the vehicle, such as a fence, a wall, a plant, etc. It can be considered that the fixed obstacle is fixed and does not move under the influence of the outside world. The preset position condition includes but is not limited to that the distance between the obstacle and the vehicle is less than a preset distance threshold, the obstacle completely blocks the vehicle in the target direction, etc. The direction of the obstacle can be divided based on the vehicle, in the direction corresponding to the image channel, such as the front of the vehicle, the rear of the vehicle, the left side of the vehicle, the right side of the vehicle, etc. For example, the obstacle is in the front of the vehicle, the rear of the vehicle, the left side of the vehicle, etc. The closing of the image channel includes stopping the image acquisition and processing in the target direction, that is, closing the hardware and / or software corresponding to the image channel, reducing the power consumption.

[0065] For example, the closing / opening of the image channel in the target direction does not affect the work of the image channel in other directions, so that the image channel in other directions except the target direction can continue to acquire the initial image, monitor the suspicious target, and capture the potential threat. For example, the closing / opening of the image channel does not affect the work of other sensing channels in the sentinel mode, and the sensing channel includes but is not limited to a laser radar channel, an infrared sensor channel, etc. The laser radar channel is composed of a laser emitter, a scanning mechanism, a photoelectric detector, a processing unit, etc., and can provide three-dimensional environmental information outside the vehicle. The infrared sensor channel is composed of a thermal imager and a signal processing unit, etc., and can make up for the deficiency of the visible light camera in the lightless environment.

[0066] By using the technical solution provided in the step, the environmental information obtained in step S102 is used to make a decision on whether to close the image channel in a certain direction. Specifically, in the case where it is detected that there is an obstacle outside the vehicle in the environmental information, if it is determined that the position relationship between the obstacle and the vehicle in at least one target direction meets the preset position condition, it means that the vehicle is safe (in the case of no risk or extremely low risk) in the target direction, and therefore the image channel in the target direction can be closed, that is, the hardware (such as a camera sensor, etc.) of the image channel is stopped from working, and the software (such as environmental information analysis, obstacle detection, etc.) of the image channel is also stopped from working, thereby saving the electric energy from the source and effectively prolonging the time length of the vehicle parking monitoring.

[0067] In one implementation manner, when the sentinel mode is started, the initial image is acquired, and the environmental information is obtained based on the initial image, and then it is timely determined whether the obstacle can be detected in the environmental information. If the obstacle is detected, and the position relationship between the obstacle and the vehicle in the target direction meets the preset position condition, the image channel corresponding to the target direction is quickly closed, and the opening time length of the image channel is reduced to the minimum, and the power consumption of the hardware and / or software corresponding to the image channel is reduced.

[0068] In an implementation, in an extreme case, the obstacle has a possibility of moving due to external influence, for example, the obstacle is a fence or other object that can be moved. Therefore, in the case that the image channel in the target direction is closed, the interval is preset for a time length, the image channel is opened, and an initial image is acquired. It is determined whether the direction corresponding to the image channel still has an obstacle. In the case that an obstacle is detected and the position relationship between the obstacle and the vehicle in the direction corresponding to the image channel satisfies a preset position condition, the image channel is closed again. In this way, it is ensured that the vehicle is safe in the direction, and the accuracy of the sentry mode monitoring is improved.

[0069] In an implementation, the method further includes detecting that the type of the obstacle is a preset fixed object type. The preset fixed object type includes, but is not limited to, a fence, a wall, a green plant, and the like. It can be considered that the height of the preset fixed object type is sufficient to block a person from climbing over, and the person cannot pose a threat to the vehicle above the obstacle. Therefore, in the case that the position relationship between the obstacle and the vehicle in at least one target direction satisfies a preset position condition, it can be considered that a person cannot enter the gap between the obstacle and the vehicle, and thus the vehicle is safe in the target direction, and the image channel corresponding to the target direction can be closed.

[0070] In the technical solution of the embodiments of the present disclosure, in the case that the sentry mode is opened, initial images corresponding to each image channel are acquired, and environmental information of the vehicle is analyzed, and an obstacle existing outside the vehicle is identified. In the case that the position relationship between the obstacle and the vehicle in a specific target direction satisfies a preset position condition, the image channel corresponding to the target direction is closed. This way effectively avoids continuous image acquisition and processing in a direction without risk or with extremely low risk, significantly reduces power consumption, prolongs the monitoring time of the vehicle in a parked state, and improves the practicality and user experience of the sentry mode.

[0071] Figure 4A A schematic diagram of a position relationship between an obstacle and a vehicle is provided for an example embodiment of the present disclosure; Figure 4B A schematic diagram of a position relationship between an obstacle and a vehicle is provided for another example embodiment of the present disclosure.

[0072] In an example embodiment, the detection of the obstacle outside the vehicle in the environmental information includes a plurality, referring to Figure 4A, the left side and the front of the vehicle D each have an obstacle, the target direction of the obstacle A is the front of the vehicle D, and the direction of the obstacle B is the left side of the vehicle. When determining the positional relationship between the obstacle A and the vehicle D, the positional relationship between the obstacle A and the vehicle D in the front of the vehicle is determined. The positional relationship between the obstacle B and the vehicle D in the left side of the vehicle is determined, and then it can be judged whether the positional relationship meets the preset position condition. For example, the projection of the obstacle A in the front of the vehicle completely covers the vehicle head, and it can be considered that the obstacle A completely blocks the vehicle D in the front of the vehicle. It can be considered that the person cannot enter the gap between the obstacle A and the vehicle D when it is detected that the distance between the obstacle A and the vehicle D is less than the preset distance threshold. In the case that the obstacle A completely blocks the vehicle D and the person cannot pass through the gap between the obstacle A and the vehicle D, it can be considered that there is no threat in the front of the vehicle, and in this case, the positional relationship between the obstacle A and the vehicle D in the front of the vehicle meets the preset position condition, and the image channel in the front of the vehicle can be closed. Similarly, the positional relationship between the obstacle B and the vehicle D in the left side of the vehicle does not meet the preset position condition, and the image channel in the left side of the vehicle is not closed.

[0073] In another exemplary embodiment, it is detected that there is one obstacle outside the vehicle in the environment, referring to Figure 4B , the obstacle C is in the oblique front of the vehicle D, for example, the directions corresponding to each image channel include the front of the vehicle, the rear of the vehicle, the left side of the vehicle and the right side of the vehicle, at this time, the initial image corresponding to the image channel in the front of the vehicle and the left side of the vehicle is part of the obstacle C, therefore, the target direction corresponding to the obstacle C includes two, the front of the vehicle and the left side of the vehicle. It is respectively judged whether the positional relationship between the obstacle C and the vehicle D in the front of the vehicle and the left side of the vehicle meets the preset position condition, in the case that the positional relationship between the obstacle C and the vehicle D in any target direction meets the preset position condition, the image channel corresponding to the target direction is closed. For example, the projection of the obstacle C in the front of the vehicle completely covers the vehicle head, and it can be considered that the obstacle C completely blocks the vehicle D in the front of the vehicle. It can be considered that the gap between the obstacle C and the vehicle D is available for the person to enter, and then the parts such as the vehicle head are threatened, when it is detected that the distance between the obstacle C and the vehicle D in the front of the vehicle is greater than the preset distance threshold. Based on the above information, it is known that the obstacle C completely blocks the vehicle D, but the distance between the obstacle C and the vehicle D is greater than the preset distance threshold, and therefore the positional relationship between the obstacle C and the vehicle D does not meet the preset position condition, and the image channel in the front of the vehicle is not closed.

[0074] According to the environmental information, whether there is an obstacle outside the vehicle is detected, and in the case that there is an obstacle, the positional relationship between the obstacle and the vehicle in at least one target direction corresponding to the obstacle is obtained, which can be realized in the following exemplary manner:

[0075] In one possible implementation, based on environmental information, it is determined whether there are obstacles in the directions corresponding to each image channel. Then, if there are obstacles in at least one target direction, the target direction of the obstacle and the positional relationship between the obstacle and the vehicle in the target direction are determined. In this implementation, the presence of obstacles in the direction corresponding to each image channel is checked separately. If there are obstacles in one or more target directions, the positional information of the obstacle and the vehicle in the target direction is further analyzed, ensuring the independence of each direction. The failure of obstacle detection in a single direction (such as due to unqualified initial image quality) will not affect the analysis of other directions.

[0076] In another possible implementation, the system detects whether there are obstacles outside the vehicle in the environmental information. If obstacles are detected, the target direction of the obstacle and its spatial relationship with the vehicle are determined based on the environmental information. In this implementation, the surrounding environment is first analyzed as a whole to determine whether there are obstacles outside the vehicle in one go. If obstacles exist, the environmental information is used to accurately determine one or more target directions of the obstacle. For each target direction, the spatial relationship between the obstacle and the vehicle is determined. Global obstacle detection is completed in one go. For complex obstacles spanning multiple directions, this implementation can directly identify the overall outline of the obstacle from a global perspective and accurately determine its direction and spatial relationship with the vehicle.

[0077] Figure 5 This is a flowchart illustrating a control method for starting and stopping in sentry mode, provided in another exemplary embodiment of this disclosure.

[0078] like Figure 5 As shown, in one implementation, after step S102, the method further includes the following steps:

[0079] Step S104: If an obstacle is detected outside the vehicle in the environmental information and the positional relationship between the obstacle and the vehicle in the target direction does not meet the preset positional conditions, determine the initial image corresponding to the target direction.

[0080] Using the technical solution provided in this step, an obstacle is detected outside the vehicle in the environmental information. However, the positional relationship between the obstacle and the vehicle in the target direction does not meet the preset positional conditions. At this time, the initial image corresponding to the target direction is determined so that the image channel of the target direction corresponding to the initial image can be closed in the future.

[0081] Step S105: In response to the image quality of the initial image being unqualified, the image channel corresponding to the target direction of the initial image is closed.

[0082] The image quality refers to the quality of the initial image itself, such as definition, and the unqualified image quality can refer to that the image cannot meet the basic quality standard required for effective monitoring and analysis, for example, image blur, image overexposure, image underexposure, large-area occlusion of the image by a cover (for example, most areas of the image display black or the same or similar color tone), and the like.

[0083] By using the technical solution provided in this step, when it is detected that the image quality of the initial image is unqualified, it can be considered that the image quality of the initial image is poor and cannot be used for effective monitoring and identification analysis. Therefore, the image channel corresponding to the target direction of the initial image is closed, and the power consumption is reduced.

[0084] The technical solution of the embodiments of the present disclosure detects that there is an obstacle outside the vehicle in the environmental information, but the positional relationship between the obstacle and the vehicle in the target direction does not meet the preset positional condition. At this time, the vehicle may be at risk in the target direction, and further checks whether the quality of the initial image corresponding to the target direction is qualified. If the quality of the initial image is unqualified, it can be considered that effective identification analysis cannot be performed according to the initial image. Therefore, the image channel corresponding to the initial image can be closed, and the continuous invalid analysis of the initial image with low quality and no value is avoided, and the calculation resources and the electric energy are saved.

[0085] In one implementation manner, in a case where it is detected that there is an obstacle outside the vehicle in the environmental information, and the positional relationship between the obstacle and the vehicle in the target direction does not meet the preset positional condition, the initial image corresponding to the target direction is determined. If the image quality of the initial image is qualified, the image channel corresponding to the target direction of the initial image remains in an open state, and the trend of the target direction is continuously monitored, so as to discover the danger in time.

[0086] In an implementation, when the sentry mode is enabled, the image quality of the initial image is determined to be qualified, and the image channel corresponding to the initial image remains in an enabled state. Due to weather changes or other reasons, at a certain time node, the initial image captured by the camera corresponding to the image channel has poor image quality, which cannot effectively monitor and identify. For example, 10 minutes after the image channel corresponding to the initial image remains in the enabled state, snow causes the camera to be largely blocked, and the like. Therefore, a first preset time length can be set. In the case where the image channel corresponding to the initial image remains in the enabled state, an initial image at a current time is acquired every interval of the first preset time length, and the image quality of the initial image is detected, so that whether the image quality of the latest initial image is qualified can be found in time, and in the case where the image quality of the initial image is not qualified, the image channel corresponding to the initial image is closed, unnecessary power consumption is reduced, and the effectiveness of sentry mode monitoring is improved. Similarly, after the image channel in the direction corresponding to the initial image is closed, a second preset time length can be set, and the image channel is re-enabled every interval of the second preset time length, so as to acquire the initial image at the current time through the image channel, and detect the image quality of the initial image, and then determine whether the image channel needs to be closed based on the image quality.

[0087] Figure 6 FIG. 1 is a flow diagram of a control method for enabling and disabling a sentry mode according to another example embodiment of the present disclosure.

[0088] As shown in FIG. 1, in an implementation, after step S102, the method further includes the following steps: Figure 6

[0089] Step S106, in the case where no obstacle is detected outside the vehicle in the environmental information, determining the initial image corresponding to each direction outside the vehicle.

[0090] The case where no obstacle is detected outside the vehicle can mean that there is no obstacle outside the vehicle, or that the environmental information obtained from each initial image cannot identify whether there is an obstacle outside the vehicle, such as the case where the initial image is blurred, and the like.

[0091] The technical solution provided in this step, after analyzing the environmental information, if it is determined that no obstacle is detected outside the vehicle, further acquiring and analyzing the initial image of each direction outside the vehicle, provides a data basis for subsequent image quality analysis, and ensures the effectiveness of the sentry mode in each direction.

[0092] Step S107, in response to the image quality of at least one initial image being unqualified, closing the image channel in the direction corresponding to the initial image.

[0093] ​Using the technical solution provided in this step, if the quality of the initial image in a certain direction is unqualified, it can be assumed that the initial image cannot effectively support the sensing function of the sentinel mode. Therefore, the image channel in the direction corresponding to the initial image is turned off to reduce unnecessary image acquisition and processing power consumption.

[0094] In this embodiment of the disclosure, if no obstacles are detected outside the vehicle in the environmental information, it can be determined whether there is an image channel that can be turned off based on the image quality of the initial image. If the image quality of the initial image is unqualified, the image channel corresponding to the direction of the initial image is turned off. By turning off invalid image channels, power waste in sentry mode is avoided.

[0095] In one possible implementation, if no obstacles are detected outside the vehicle in the environmental information, an initial image is determined for each direction outside the vehicle. If the image quality of the initial image is qualified, the image channel for the target direction corresponding to the initial image remains open to continuously monitor the movement in the target direction in order to detect dangers in a timely manner.

[0096] Figure 7 This is a schematic flowchart illustrating an exemplary embodiment of the present disclosure of a method for determining image quality.

[0097] like Figure 7 As shown, in one implementation, steps S108-S110 are included before step S107. It should be noted that the image quality determination method of this embodiment can also be implemented before step S105. The specific solutions for steps S108-S110 are as follows:

[0098] Step S108: Obtain the image quality category of multiple pixels in the preset central region of the initial image.

[0099] The preset center region is a predefined central area in the initial image, representing the core of the initial image. The image quality category of the preset center region can be used as the image quality category of the initial image. The size and shape of the preset center region can be set according to the actual situation. For example, if the camera capturing the initial image is a fisheye camera, the preset center region can be set to a circle. The image quality category can refer to pixel-level quality categories, such as low light due to insufficient ambient light, water droplets left by the camera, blurring due to focus failure or camera shake, camera obstruction by mud, snow, or other objects, overexposure due to strong light, and normal quality.

[0100] Using the technical solution provided in this step, the image quality category of multiple pixels included in the preset central region of the initial image is obtained in order to determine the overall image quality of the preset central region.

[0101] Step S109, determining the target pixel point of the preset image quality category from the plurality of pixel points based on the image quality categories of the plurality of pixel points.

[0102] The preset image quality category is a pre-set image quality category, and can include non-normal categories such as dark light, water stains, blur, occlusion, and overexposure.

[0103] The technical solution provided in this step filters the target pixel point belonging to the preset image quality category from the pixel points in the preset central region. The target pixel point affects the image quality, and therefore can be used as a basis for judging whether the image quality is qualified.

[0104] Step S110, in response to the ratio of the number of target pixel points to the total number of pixel points in the preset central region being greater than a preset threshold, determining that the image quality of the initial image is unqualified.

[0105] The preset threshold can be set according to actual conditions, and the present embodiment does not make specific limitations on the preset threshold. For example, the preset threshold can be 0.8.

[0106] The technical solution provided in this step shows that if the ratio of the number of target pixel points to the total number of pixel points in the preset central region is greater than the preset threshold, it indicates that most of the pixel points in the preset central region are in a non-normal category, and it can be considered that the image quality of the preset central region is unqualified, and therefore it can be determined that the image quality of the initial image is unqualified. Converting the unqualified image quality into a calculable ratio index makes the accuracy of closing the image channel corresponding to the initial image higher based on the unqualified image quality of the initial image.

[0107] The technical solution of the present embodiment quantitatively judges whether the initial image is qualified through pixel-level image quality analysis. Specifically, the image quality categories of the pixel points of the initial image in the preset central region are obtained, and the target pixel point belonging to the preset image quality category is filtered out. Through comparison of the number of target pixel points and the total number of pixel points in the preset central region, the judgment standard of image quality is unified and repeatable, and the reliability of decision-making is improved. Based on the pixel-level image quality classification, multiple complex image quality problems such as occlusion, blur, overexposure, and dark light can be simultaneously addressed, and the corresponding image channel that cannot be effectively monitored and analyzed due to unqualified image quality can be accurately identified.

[0108] Figure 8 is a flowchart of an image quality category acquisition method provided by an exemplary embodiment of the present disclosure.

[0109] As shown in Figure 8 in an implementation manner, step S108 can include the following steps:

[0110] Step S1080, performing image quality classification on the initial image by the trained image quality classification model to obtain the image quality category of each pixel point in the initial image.

[0111] The image quality classification model can be trained by first sample images of various image quality categories and image quality category labels corresponding to the first sample images. The image quality category labels include, but are not limited to, a dark light label, a water stain label, an occlusion label, a blur label, and a normal label, etc. This training method can ensure that the classification result of the image quality classification model has high accuracy and reliability. The image quality classification model can perform pixel-level image quality classification on the input image to obtain the image quality category of each pixel point in the input image. The image quality classification model includes, but is not limited to, a fully convolutional network (FCN) and a recurrent neural network (RNN).

[0112] By adopting the technical solution provided in this step, the image quality classification model is trained to perform image quality classification on the initial image to obtain the image quality category of each pixel point in the initial image, so as to obtain the image quality category of the pixel point in the preset central region in the subsequent step.

[0113] Step S1081, obtaining the image quality categories of the plurality of pixel points in the preset central region of the initial image.

[0114] By adopting the technical solution provided in this step, the image quality categories of the plurality of pixel points in the preset central region are obtained from the image quality categories of each pixel point in the initial image.

[0115] The technical solution of the embodiment of the present disclosure performs pixel-by-pixel image quality classification on the entire initial image by the trained image quality classification model to obtain the image quality category of each pixel point. By simply reading the image quality category of each pixel point in the preset central region, the quality condition of the core region that needs to be focused on can be quickly locked, and the efficiency and comprehensiveness of the processing are improved.

[0116] Figure 9 A schematic diagram of the position relationship between the obstacle and the vehicle is provided for another exemplary embodiment of the present disclosure.

[0117] In an exemplary embodiment, the preset position condition includes whether the obstacle completely occludes the vehicle in the target direction, and whether the distance between the obstacle and the vehicle in the target direction is less than a preset distance threshold.

[0118] The preset position condition includes two criteria, through the judgment of double physical space relations, the adaptive start-stop of the image channel of the sentinel mode is realized, when one side of the vehicle is completely blocked by the fixed obstacle and the distance is very close, it can be considered that there is no threat in this direction, the continuous detection has limited significance, but the traditional system still keeps running, resulting in waste of power consumption, while the preset position relationship in the example of the disclosure limits whether the obstacle completely blocks the vehicle and whether the distance between the obstacle and the vehicle is less than the preset distance threshold. When the preset position relationship between the obstacle and the vehicle is satisfied, the image channel in the direction of the obstacle can be closed, and the waste of power consumption is reduced.

[0119] As Figure 9 (1) in Figure 9 (3) shown, the obstacle E is located in front of the vehicle D, Figure 9 (1) in shows the case that the obstacle E completely blocks the vehicle D, Figure 9 (2) in and Figure 9 (3) in shows the case that the obstacle E does not completely block the vehicle D.

[0120] In an exemplary embodiment, the distance between the obstacle and the vehicle can be the longest distance between the obstacle and the vehicle, or the average of the distances between the multiple preset positions of the vehicle and the obstacle, for example, if the obstacle E is parallel to the vehicle, the vertical distance between the obstacle E and the vehicle can be determined, and for example, in each direction, the distance measurement points of the vehicle are set in advance, if the target direction is in front of the vehicle, the center points of the two headlamps and the license plate are set as the distance measurement points, the distances between the center points of the two headlamps, the license plate and the obstacle are obtained respectively, and then the average of the three distance values is taken as the distance between the obstacle and the vehicle. Of course, the example is only an optional example, and does not limit the technical solutions of the disclosure.

[0121] Exemplarily, in front of the vehicle, whether the obstacle completely blocks the vehicle is related to the length and position of the obstacle, as Figure 9As shown in (2) of FIG. 6, in the case where the projection length of the obstacle E in front of the vehicle is greater than the width of the vehicle D, whether the obstacle E completely blocks the vehicle D can be determined according to the relative position between the obstacle E and the vehicle D. For example, a coordinate system is established with the vehicle D as the reference, and the position relationship between the vehicle D and the obstacle E is determined according to the horizontal and vertical coordinates. In front of the vehicle, the coordinates (X21, Y21) and (X22, Y22) of the two ends of the obstacle E are obtained, and the value of |X22-X21| is the projection length of the obstacle in front of the vehicle. The coordinates (x11, y11) and (x12, y12) of the two ends of the vehicle D are obtained, and the value of |x12-x11| is the width of the vehicle D. Therefore, if |X22-X21|>|x12-x11|, the projection length of the obstacle E in front of the vehicle is greater than the width of the vehicle D. If X22>x12 and X21>x11, it can be considered that the position of the obstacle E is in the front right of the vehicle D, and the obstacle E does not completely block the vehicle D. As shown in (2) of FIG. 6, the coordinates of the two ends of the vehicle D can refer to the position with the widest body width, for example, the outermost side of the outside rearview mirror on both sides of the vehicle D. Figure 9 As shown in (3) of FIG. 6, in the case where the projection length of the obstacle E in front of the vehicle is less than or equal to the width of the vehicle D, it can be determined that the obstacle E cannot completely block the vehicle D.

[0122] For example, for the position relationship between the obstacle and the vehicle, the laser radar positioning method can also be used. Specifically, laser radars are arranged around the vehicle. The distance, direction, and reflection intensity of the obstacle in front of the vehicle are measured by the laser radars to generate a three-dimensional point cloud model, so as to obtain the specific position of the obstacle in space, the relative position between the obstacle and the vehicle, and the outer contour of the obstacle, and further obtain whether the obstacle completely blocks the vehicle, the distance between the obstacle and the vehicle, and the like. Thus, the position relationship between the obstacle and the vehicle can be known.

[0123] In an example embodiment, in combination with Figure 4B In the case where the position of the obstacle C and the vehicle D has a certain angle, the obstacle C is located in the oblique front of the vehicle D, and the target direction corresponding to the obstacle C includes the front of the vehicle and the left side of the vehicle. Therefore, when determining whether the obstacle C in the direction completely blocks the vehicle D, the determination of two directions is included, that is, determining whether the obstacle C completely blocks the vehicle D in the front of the vehicle, and determining whether the obstacle C completely blocks the vehicle D in the left side of the vehicle. Figure 4B As shown in (2) of FIG. 6, in the case where the projection length of the obstacle E in front of the vehicle is greater than the width of the vehicle D, whether the obstacle E completely blocks the vehicle D can be determined according to the relative position between the obstacle E and the vehicle D. For example, a coordinate system is established with the vehicle D as the reference, and the position relationship between the vehicle D and the obstacle E is determined according to the horizontal and vertical coordinates. In front of the vehicle, the coordinates (X21, Y21) and (X22, Y22) of the two ends of the obstacle E are obtained, and the value of |X22-X21| is the projection length of the obstacle in front of the vehicle. The coordinates (x11, y11) and (x12, y12) of the two ends of the vehicle D are obtained, and the value of |x12-x11| is the width of the vehicle D. Therefore, if |X22-X21|>|x12-x11|, the projection length of the obstacle E in front of the vehicle is greater than the width of the vehicle D. If X22>x12 and X21>x11, it can be considered that the position of the obstacle E is in the front right of the vehicle D, and the obstacle E does not completely block the vehicle D. As shown in (2) of FIG. 6, the coordinates of the two ends of the vehicle D can refer to the position with the widest body width, for example, the outermost side of the outside rearview mirror on both sides of the vehicle D.

[0124] For example, in front of the vehicle, the projection length of the obstacle in front of the vehicle is obtained, and if the projection length is greater than the width of the vehicle, it is determined according to the relative position between the obstacle and the vehicle whether the obstacle completely blocks the vehicle. The determination of whether the obstacle completely blocks the vehicle according to the relative position between the obstacle and the vehicle can refer to the foregoing embodiments, which will not be described here. Of course, if the projection length of the obstacle is less than or equal to the width of the vehicle, it can be determined that the obstacle cannot completely block the vehicle.

[0125] In an implementation manner, the length of the obstacle is determined according to the environmental information obtained from each initial image, which can also be understood as the length of the obstacle within the range that can be captured by the camera of the vehicle. For example, the obstacle is a wall, and the camera of the vehicle can only capture part of the wall. The length of the obstacle involved in the embodiments of the present disclosure is only the length of the captured part of the wall.

[0126] Figure 10 A schematic diagram of the length of an obstacle provided by an exemplary embodiment of the present disclosure.

[0127] In order to further illustrate the size of the obstacle in the embodiments of the present disclosure, for example, as shown in Figure 10 In the capturing range of each image channel corresponding to the camera, that is, the fan-shaped area 11 to the fan-shaped area 14, the length of the obstacle F is L2, and the complete length of the non-obstacle is L3. The length L2 is used to determine the size relationship between the obstacle and the width of the vehicle.

[0128] Figure 11 A flowchart of a method for determining the relative position between an obstacle and a vehicle provided by an exemplary embodiment of the present disclosure.

[0129] As shown in Figure 11 In an implementation manner, after step S102, the method further includes the following steps:

[0130] In step S111, the global bird's eye view corresponding to the environmental information is subjected to obstacle detection by using the trained obstacle detection model, and a detection result is obtained.

[0131] The global bird's eye view is an initial image stitching of multiple image channels, for example, the initial images of the surround-view fisheye cameras in front, rear, left and right of the vehicle are acquired, the initial images are de-distorted, a homography matrix (H) is constructed based on a camera imaging model, image pixel coordinates (u, v) are mapped to ground plane coordinates (X, Y) = H-1(u, v) through inverse perspective transformation, and thus local bird's eye views in various directions are obtained. The projection results (i.e. local bird's eye views) in various directions are geometrically aligned and color fused in the overlapping area to generate a global bird's eye view centered on the vehicle. The global bird's eye view can be used as a unified bird's eye space representation and provides a bird's eye view coordinate system.

[0132] For example, the homography matrix (H) is as follows in formula (1):

[0133]

[0134] wherein s is a scale factor, K is a camera intrinsic matrix, R is a rotation matrix, and t is a translation vector.

[0135] The obstacle detection model can be trained by second sample images of various obstacles and obstacle labels corresponding to the second sample images. The obstacle labels include but are not limited to walls, stone piers, fences, and green plants. The detection results can include whether an obstacle is detected in the global bird's eye view, a target direction of the obstacle, position information of the obstacle, and of course, the type of the obstacle. The obstacle detection model includes but is not limited to a convolutional neural network model.

[0136] In step S111, the bird's eye view provides a God's eye view, eliminates the overlapping problem caused by different camera perspectives, and makes the obstacle detection and subsequent geometric relationship analysis more accurate and intuitive. By using the trained obstacle detection model, the global bird's eye view corresponding to the environmental information is subjected to obstacle detection, various types of obstacles can be automatically and real-timely identified, manual setting of rules is not required, and the adaptability is high.

[0137] In step S112, in the case that the detection result includes an obstacle existing outside the vehicle, the target direction of the obstacle and the position information of the obstacle are determined based on the detection result.

[0138] The target direction of the obstacle includes various cases, as shown in Figure 4A , Figure 4B and Figure 9 The target direction of the obstacle C includes two, i.e. the front of the vehicle and the left side of the vehicle, and the target direction of the obstacle E includes only one, i.e. the front of the vehicle.

[0139] In the implementation manner of the step, in the case that the detection result includes that there is an obstacle outside the vehicle, the detected obstacle is associated with the direction corresponding to the specific image channel, and the specific position of the obstacle in the bird's eye view coordinate system is obtained.

[0140] The step specifies the specific object of risk analysis, for example, the obstacle is identified on the left side of the vehicle, and all subsequent analyses will focus on the left side, so that the decision has clear pertinence.

[0141] In an implementation manner of the present disclosure, the detection result further includes an obstacle type, and if the obstacle type is a preset fixed object type, it can be considered that the height of the obstacle is sufficient to block the person from climbing over. Therefore, the target direction where the obstacle is located and the position information of the obstacle are determined, so as to subsequently judge the positional relationship between the vehicle and the obstacle.

[0142] In step S113, the position information of the vehicle, the size information of the vehicle and the size information of the obstacle are determined according to the global bird's eye view.

[0143] The size information of the vehicle can include the length of the vehicle and the width of the vehicle. The size information of the obstacle refers to the size information of the side close to the vehicle, which can be seen from Figure 4A 、 Figure 4B 、 Figure 9 and Figure 10 . It should be noted that the size information of the obstacle in the embodiment of the present disclosure is determined according to the global bird's eye view obtained from each initial image.

[0144] In the implementation manner of the step, the position information of the vehicle, the size information of the vehicle and the size information of the obstacle can be obtained in the same bird's eye view coordinate system. The global bird's eye view eliminates the problems such as perspective distortion, and the contour, size and relative position relationship between the vehicle and the obstacle can be quantitatively processed, so that the subsequent judgment is more accurate.

[0145] In step S114, whether the obstacle completely blocks the vehicle in the target direction is judged according to the position information of the obstacle, the size information of the obstacle, the position information of the vehicle and the size information of the vehicle.

[0146] In the implementation manner of the step, the relative position relationship between the vehicle and the obstacle can be obtained according to the position information of the obstacle and the position information of the vehicle. According to the relative position relationship between the obstacle and the vehicle, the size information of the obstacle and the size information of the vehicle, it can be determined whether the obstacle completely blocks the vehicle in the target direction.

[0147] In an optional embodiment, referring to Figure 9In (1), whether the obstacle E completely blocks the vehicle D in the bird's eye view coordinate system can be obtained according to the coordinate information of the obstacle E and the vehicle D, for example, whether the obstacle E completely blocks the vehicle D is determined by the longitudinal coordinates of the obstacle E and the vehicle D, X11 < x11, X12 > x12, which means that the obstacle E completely blocks the vehicle D. Of course, this is only an example and does not limit the technical solutions of the embodiments of the present disclosure.

[0148] In step S115, whether the distance between the obstacle and the vehicle in the target direction is less than a preset distance threshold is determined according to the position information of the obstacle and the position information of the vehicle.

[0149] The preset distance threshold provides a configurable safety parameter, which can be adjusted according to different vehicle models or user preferences to balance the energy saving effect and safety redundancy. The present disclosure does not make specific limitations on the setting of the preset distance threshold.

[0150] According to the position information of the obstacle and the position information of the vehicle, whether the distance between the obstacle and the vehicle in the target direction is less than a preset distance threshold is determined, and the situation that the obstacle is close to the vehicle is identified. The physical space close to the vehicle greatly limits the activity possibility of potential threats.

[0151] In the embodiments of the present disclosure, the global bird's eye view corresponding to the environmental information is detected by the trained obstacle detection model to obtain a detection result. When the detection result includes an obstacle existing outside the vehicle, the target direction of the obstacle and the position information of the obstacle are obtained. Then, the position information of the vehicle, the size information of the vehicle and the size information of the obstacle are obtained according to the global bird's eye view. Then, whether the obstacle completely blocks the vehicle in the target direction and whether the distance between the obstacle and the vehicle is less than a preset distance threshold are determined based on the obtained information. The double determination mechanism is used to determine whether the vehicle is safe in the target direction, and then the image channel corresponding to the target direction can be closed in the case that the vehicle is safe to reduce power waste.

[0152] The scheme of the present disclosure is not limited to the embodiments mentioned above.

[0153] Exemplary device

[0154] The control method of the sentinel mode start-stop provided by the embodiments of the present disclosure is introduced. It can be understood that the control device of the sentinel mode start-stop can include corresponding hardware and software for realizing the functions of the control method of the sentinel mode start-stop.

[0155] Those skilled in the art should easily realize that the steps of the control method of the sentinel mode start-stop described in combination with the embodiments of the present disclosure can be implemented in the form of hardware, or in the form of software driving hardware in combination. Whether a certain function is implemented in hardware or in the form of software driving hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present disclosure.

[0156] Figure 12 FIG. 1 is a structural schematic diagram of a control device for a sentinel mode start-stop according to an example embodiment of the present disclosure.

[0157] As shown in FIG. 1, in one embodiment, the control device 100 for the sentinel mode start-stop includes: Figure 12

[0158] an image acquisition component 110 including a plurality of cameras 111 outside the vehicle, the cameras 111 being configured to acquire initial images outside the vehicle through the corresponding cameras 111 of the image channels when the sentinel mode is started; and one or more processors 120 configured to execute instructions stored in a memory to obtain environmental information of the vehicle according to the initial images corresponding to each image channel, and to close the image channel corresponding to at least one target direction in response to detecting that there is an obstacle outside the vehicle in the environmental information and that the positional relationship between the obstacle and the vehicle in the at least one target direction satisfies a preset positional condition.

[0159] In the technical solution of the embodiments of the present disclosure, the image acquisition component 110 acquires the initial images corresponding to each image channel when the sentinel mode is started, and the one or more processors 120 analyze the environmental information of the vehicle to identify the obstacle existing outside the vehicle. When the positional relationship between the obstacle and the vehicle in a specific target direction satisfies the preset positional condition, the image channel corresponding to the target direction is closed. This way effectively avoids continuous image acquisition and processing in the direction with no risk or extremely low risk, significantly reduces power consumption, prolongs the monitoring time of the vehicle in the parked state, and improves the practicality and user experience of the sentinel mode.

[0160] In one example embodiment, after obtaining the environmental information of the vehicle according to the initial images corresponding to each image channel, the processor 120 is configured to execute instructions stored in the memory to perform the following operations:

[0161] ​In a case where it is detected that there is an obstacle outside the vehicle in the environmental information, and a position relationship between the obstacle and the vehicle in the target direction does not satisfy a preset position condition, an initial image corresponding to the target direction is determined; in response to an image quality of the initial image being unqualified, an image channel corresponding to the target direction in the initial image is closed.

[0162] In one example embodiment, after obtaining the environmental information of the vehicle according to the initial image corresponding to each image channel, the processor 120 is configured to execute the instructions stored in the memory to perform the following operations:

[0163] In a case where no obstacle is detected outside the vehicle in the environmental information, an initial image corresponding to each direction outside the vehicle is determined; in response to an image quality of at least one initial image being unqualified, an image channel corresponding to the direction of the initial image is closed.

[0164] In one example embodiment, before closing the image channel corresponding to the direction of the initial image in response to the image quality of the at least one initial image being unqualified, the processor 120 is configured to execute the instructions stored in the memory to perform the following operations:

[0165] Obtaining image quality categories of a plurality of pixel points in a preset central region of the initial image; determining a target pixel point of a preset image quality category from the plurality of pixel points based on the image quality categories of the plurality of pixel points; in response to a ratio of a number of the target pixel points to a total number of pixel points in the preset central region being greater than a preset threshold, determining that the image quality of the initial image is unqualified.

[0166] In one example embodiment, the processor 120 is configured to execute the instructions stored in the memory to perform the following operations:

[0167] Classifying the image quality of the initial image by using the trained image quality classification model to obtain an image quality category of each pixel point in the initial image; and obtaining image quality categories of a plurality of pixel points in a preset central region of the initial image.

[0168] In one example embodiment, after obtaining the environmental information of the vehicle according to the initial image corresponding to each image channel, the processor 120 is configured to execute the instructions stored in the memory to perform the following operations:

[0169] The trained obstacle detection model is used to detect obstacles in the global bird's eye view corresponding to the environment information, to obtain a detection result; in a case where the detection result includes an obstacle existing outside the vehicle, the target direction where the obstacle is located and position information of the obstacle are determined based on the detection result; position information of the vehicle, size information of the vehicle, and size information of the obstacle are determined according to the global bird's eye view; whether the obstacle completely blocks the vehicle in the target direction is determined according to the position information of the obstacle, the size information of the obstacle, the position information of the vehicle, and the size information of the vehicle; and whether the distance between the obstacle and the vehicle in the target direction is less than a preset distance threshold is determined according to the position information of the obstacle and the position information of the vehicle.

[0170] An example electronic device

[0171] Figure 13 A structural schematic diagram of an electronic device is provided for an example embodiment of the present disclosure, including at least one processor 1310 and a memory 1320.

[0172] The processor 1310 can be a central processing unit (CPU) or other form of processing unit having data processing and / or instruction execution capabilities, and can control other components in the electronic device 1300 to perform desired functions.

[0173] The memory 1320 can include one or more computer program products, which can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may, for example, include random access memory (RAM), cache, and / or the like. Non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, and / or the like. One or more computer program instructions can be stored on the computer-readable storage media, and the processor 1310 can execute the one or more computer program instructions to implement the control method for starting and stopping the sentinel mode and / or other desired functions of various embodiments of the present disclosure described above.

[0174] In one example, the electronic device 1300 can further include an input device 1330 and an output device 1340, which are interconnected through a bus system and / or other forms of connection mechanisms (not shown).

[0175] The input device 1330 can further include, for example, a keyboard, a mouse, and / or the like.

[0176] The output device 1340 can output various information to the outside, which can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and / or the like.

[0177] Of course, in order to simplify, Figure 13 Only some of the components of the electronic device 1300 related to the present disclosure are shown in the middle, and components such as buses, input / output interfaces, and the like are omitted. In addition, the electronic device 1300 can further include any other appropriate components according to a specific application.

[0178] Exemplary computer program product and computer readable storage medium

[0179] In addition to the above method and device, embodiments of the present disclosure can also provide a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps of the control method of the start-stop of the sentinel mode of various embodiments of the present disclosure described in the "Exemplary Method" section above.

[0180] The computer program product can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, etc., and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server.

[0181] In addition, embodiments of the present disclosure can also be a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the steps of the control method of the start-stop of the sentinel mode of various embodiments of the present disclosure described in the "Exemplary Method" section above.

[0182] The computer readable storage medium can be any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium, for example, but not limited to, includes an electrical, a magnetic, an optical, an electromagnetic, an infrared, or a semiconductor system, device or apparatus, or any suitable combination of the above. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0183] The basic principles of the present disclosure are described above with reference to specific embodiments, but the advantages, benefits and effects mentioned in the present disclosure are only examples and are not considered to be mandatory for each embodiment of the present disclosure. In addition, the specific details of the above disclosure are only for the purpose of illustration and understanding, and are not considered to limit the present disclosure to the above specific details. It is necessary to implement the present disclosure.

[0184] Various modifications and changes can be made to the present disclosure by those skilled in the art without departing from the spirit and scope of the present application. Thus, it is intended that the present disclosure include all such modifications and changes as fall within the scope of the claims of the present disclosure and their equivalents.

Claims

1. A control method for starting and stopping sentry mode, comprising: When Sentinel mode is enabled, the initial image corresponding to each image channel is acquired; Based on the initial images corresponding to each image channel, obtain the vehicle's environmental information; In response to detecting an obstacle outside the vehicle in the environmental information, and the positional relationship between the obstacle and the vehicle in at least one target direction satisfies a preset positional condition, the image channel corresponding to the at least one target direction is closed.

2. The control method for starting and stopping the sentry mode according to claim 1, wherein, After obtaining the vehicle's environmental information based on the initial images corresponding to each image channel, the method further includes: If an obstacle is detected outside the vehicle in the environmental information, and the positional relationship between the obstacle and the vehicle in the target direction does not meet the preset positional conditions, an initial image corresponding to the target direction is determined. In response to the initial image having substandard image quality, the image channel corresponding to the target direction of the initial image is closed.

3. The control method for starting and stopping the sentry mode according to claim 1, wherein, After obtaining the vehicle's environmental information based on the initial images corresponding to each image channel, the method further includes: If no obstacles are detected outside the vehicle in the environmental information, an initial image corresponding to each direction outside the vehicle is determined; In response to at least one initial image having substandard image quality, the image channel corresponding to the direction of the initial image is closed.

4. The control method for starting and stopping the sentry mode according to claim 3, wherein, Before closing the image channel in the direction corresponding to at least one initial image due to image quality defects, the method further includes: Obtain the image quality category of multiple pixels in a preset central region of the initial image; Based on the image quality category of the plurality of pixels, a target pixel of a preset image quality category is determined from the plurality of pixels; If the ratio of the number of target pixels to the total number of pixels in the preset central region is greater than a preset threshold, the image quality of the initial image is determined to be unqualified.

5. The control method for starting and stopping the sentry mode according to claim 4, wherein, The process of obtaining the image quality category of multiple pixels in a preset central region of the initial image includes: The initial image is classified using a trained image quality classification model to obtain the image quality category of each pixel in the initial image. Obtain the image quality category of multiple pixels in a preset central region of the initial image.

6. The control method for starting and stopping the sentry mode according to claim 1, wherein, The preset position conditions include whether the obstacle completely obscures the vehicle in the target direction, and whether the distance between the obstacle and the vehicle in the target direction is less than a preset distance threshold.

7. The control method for starting and stopping the sentry mode according to claim 6, wherein, After obtaining the vehicle's environmental information based on the initial images corresponding to each image channel, the method further includes: Using a trained obstacle detection model, obstacles are detected on the global bird's-eye view corresponding to the environmental information, and the detection results are obtained. If the detection result includes the presence of an obstacle outside the vehicle, the target direction of the obstacle and the location information of the obstacle are determined based on the detection result. Based on the global bird's-eye view, determine the vehicle's location information, the vehicle's size information, and the obstacle's size information; Based on the location information of the obstacle, the size information of the obstacle, the location information of the vehicle, and the size information of the vehicle, determine whether the obstacle completely obstructs the vehicle in the target direction; Based on the location information of the obstacle and the location information of the vehicle, determine whether the distance between the obstacle and the vehicle in the target direction is less than a preset distance threshold.

8. A control device for starting and stopping a sentry mode, comprising: The image acquisition component includes multiple cameras on the exterior of the vehicle. When the cameras are configured to be activated in sentry mode, they acquire initial images of the exterior of the vehicle through the cameras corresponding to the image channels. One or more processors are configured to execute instructions stored in memory to obtain environmental information of the vehicle based on initial images corresponding to each image channel; In response to detecting an obstacle outside the vehicle in the environmental information, and the positional relationship between the obstacle and the vehicle in at least one target direction satisfies a preset positional condition, the image channel corresponding to the at least one target direction is closed.

9. An electronic device, comprising: One or more processors, and a memory; the memory storing computer instructions; the computer instructions, when executed by the processor, cause the processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 7.

Images