Intelligent protection perception method and system

Abstract

The application is suitable for the field of computers and provides an intelligent protection perception method and system. The method comprises the following steps: acquiring a first image, wherein the first image comprises an image of an intersection position; extracting target information of a moving target according to the first image, wherein the target information comprises a moving speed and target features; identifying a protection target in a first section corresponding to the intersection position, wherein the protection target comprises a parked vehicle on a roadside and a roadside portal; determining an interval monitoring area where the protection target is located, wherein the interval monitoring area at least comprises a first area and a second area distributed along a direction of the moving speed; and estimating a first time when the moving target enters the second area according to the moving speed. The technical scheme of the application embodiment can determine whether the moving target forms a potential threat to the protection target without monitoring the protection target all the time, thereby reducing the power consumption required for monitoring.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of computer, and particularly relates to an intelligent protection perception method and system. BACKGROUND

[0002] With the increasing popularity of Internet of Things applications, various Internet of Things devices access the Internet to serve various fields such as smart energy, smart home, smart manufacturing, and intelligent visual perception refers to the information of optical sensors, learning from biological visual perception mechanism, fusing computer vision processing method and related intelligent algorithm, and performing the process of perception, understanding and prediction of image information with computer operation as the core.

[0003] In the prior art, when dealing with the protection of some protection targets, the possible threat factors corresponding to the protection targets need to be continuously identified and monitored, especially for more than one protection target, so that the above prior art causes the identification and monitoring time to be relatively long, and the required power consumption of the monitoring to be relatively large. SUMMARY

[0004] The purpose of the embodiment of the present application is to provide an intelligent protection perception method and system, which aims to solve the problems raised in the above background technology.

[0005] The embodiment of the present application is implemented in this way, on the one hand, an intelligent protection perception method, the method comprises the following steps:

[0006] Obtain a first image, the first image includes an image of an intersection position;

[0007] According to the first image, target information of a moving target is extracted, the target information includes a moving speed and a target feature;

[0008] Identify a protection target in a first section corresponding to the intersection position, the protection target includes a parked vehicle on the roadside and a roadside portal;

[0009] Determine an interval monitoring area where the protection target is located, the interval monitoring area at least includes a first area and a second area distributed along the direction of the moving speed;

[0010] Estimate a first time when the moving target enters the second area according to the moving speed;

[0011] Determine a start period according to the first time, and monitor the activity of the moving target in the interval monitoring area in the start period, the start period is a period before the first time and adjacent to the first time.

[0012] As a further scheme of the present application, the target information of the moving target is extracted according to the first image, which includes:

[0013] identify an active target in the image of the intersection position;

[0014] frame the image, and determine a trend of the active target according to at least two frames of images;

[0015] extract a first feature of the active target when the trend is towards the first road section;

[0016] determine whether the first feature meets a feature condition;

[0017] if yes, determine the first feature as a target feature.

[0018] As a further scheme of the present application, the method further comprises:

[0019] identify a plurality of second road sections connected to the intersection position when the first feature does not meet the feature condition;

[0020] extract a second image containing the first feature in the plurality of second road sections;

[0021] screen a second sub-image meeting a clear condition in the second image;

[0022] identify a second feature of the active target in the second sub-image, and combine the first feature and the second feature as the target feature of the active target.

[0023] As a further scheme of the present application, the monitoring of the activity of the active target in the interval monitoring area during the starting period comprises:

[0024] issue a starting instruction to a monitoring device of a roadside portal before the starting period to start the monitoring device and feed back an initial monitoring angle;

[0025] calculate a deflection angle of the monitoring device covering the parked vehicle according to the positional relationship between the roadside portal and the parked vehicle;

[0026] calculate a total coverage angle according to the initial monitoring angle and the deflection angle, and generate a monitoring instruction during the starting period according to the total coverage angle and the target feature, so that the monitoring device monitors the active target according to the target feature.

[0027] As a further scheme of the present application, the method further comprises:

[0028] when the active target meets a preset condition according to the target feature during the starting period, send an abnormal prompt information to a first terminal bound to the parked vehicle and a second terminal bound to the roadside portal, the preset condition including that the active target is not in the intermediate area between the first area and the second area.

[0029] As a further scheme of the present application, the method further comprises:

[0030] When no operation response of any one of the first terminal and the second terminal is detected within a set time length, an alarm information is issued by the monitoring device, and the preset condition includes that the moving target is not in the intermediate region between the first region and the second region.

[0031] As a further aspect of the present application, in another aspect, an intelligent protection perception system, the system comprises:

[0032] An acquisition module is configured to acquire a first image, the first image comprising an image of an intersection position;

[0033] An extraction module is configured to extract target information of a moving target according to the first image, the target information comprising a moving speed and a target feature;

[0034] An identification module is configured to identify a protection target in a first section corresponding to the intersection position, the protection target comprising a parked vehicle on a roadside and a roadside portal;

[0035] A detection region determination module is configured to determine an interval monitoring region in which the protection target is located, the interval monitoring region comprising at least a first region and a second region distributed along a direction of the moving speed;

[0036] An estimation module is configured to estimate a first time at which the moving target enters the second region according to the moving speed;

[0037] A timing start monitoring is configured to determine a start period according to the first time, and monitor a moving condition of the moving target in the interval monitoring region in the start period, the start period being a period before and adjacent to the first time.

[0038] The extraction module specifically comprises:

[0039] A target identification unit is configured to identify the moving target in the image of the intersection position;

[0040] A frame dividing and determining unit is configured to divide frames of the image, and determine a moving trend of the moving target according to at least two frames of images;

[0041] An extraction unit is configured to extract a first feature of the moving target when the moving trend is towards the first section;

[0042] A judgment unit is configured to judge whether the first feature meets a feature condition;

[0043] A condition determination unit is configured to determine the first feature as the target feature if the first feature meets the feature condition.

[0044] The intelligent protection perception method and system provided by the embodiment of the application face continuous protection targets (including parked vehicles and roadside portals), determine the starting period, strictly depend on the speed of the moving target, and can determine the activity of the moving target without monitoring the moving target all the time when monitoring the protection targets in the first area and the second area, thereby reducing the power consumption required for monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 It is a main flowchart of an intelligent protection perception method.

[0046] Figure 2 It is a flowchart of extracting target information of a moving target according to the first image in the intelligent protection perception method.

[0047] Figure 3 It is a flowchart of taking the combination of the first feature and the second feature as the target feature of the moving target in the intelligent protection perception method.

[0048] Figure 4 It is a flowchart of monitoring the activity of the moving target in the interval monitoring area in the starting period in the intelligent protection perception method.

[0049] Figure 5 It is a main structure diagram of an intelligent protection perception system.

[0050] Figure 6 It is a structure diagram of an extraction module in the intelligent protection perception system. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the application more clear, the application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and do not limit the application.

[0052] The specific implementation of the application is described in detail below in combination with specific embodiments.

[0053] The intelligent protection perception method and system provided by the application solve the technical problems in the background art.

[0054] As shown in Figure 1 It is a main flowchart of an intelligent protection perception method provided by an embodiment of the application, and the intelligent protection perception method comprises:

[0055] Step S10: acquiring a first image, the first image comprising an image of an intersection position;

[0056] The first image is generally acquired at the intersection, and the acquisition position is generally located at a position opposite to the first road segment at the intersection;

[0057] Step S11: extracting target information of the moving target according to the first image, the target information including a moving speed and a target feature;

[0058] The first image is decomposed to obtain a video or an image, and the target feature can be obtained by performing feature extraction on the image. The moving speed can be roughly identified according to a path passed by the moving target in a plurality of frames of images. The moving speed is a ratio of a path length (for example, an equidistant marking line can be set at the intersection position) and a time length occupied by the plurality of frames of images, that is, the moving speed is generally an average speed, and of course, the average speed of consecutive average speeds can also be taken. In fact, the moving speed can also be collected by a common speed measuring device in the road. At this time, the speed is an instantaneous speed, and the average of a plurality of instantaneous speeds can be taken;

[0059] Step S12: identifying a protection target in the first road segment corresponding to the intersection position, the protection target including a parked vehicle at the roadside and a roadside portal;

[0060] The parked vehicle and the roadside portal are generally separated and set up. The parked vehicle generally does not stop at the front door of the roadside portal;

[0061] Step S13: determining an interval monitoring area where the protection target is located, the interval monitoring area at least including a first area and a second area distributed along a direction of the moving speed;

[0062] Generally, the protection area corresponding to the parked vehicle is determined as the first area, and the protection area corresponding to the roadside portal is determined as the second area. Of course, the reverse setting can also be used, and the present application is explained by taking the former as an example;

[0063] Step S14: estimating a first time when the moving target enters the second area according to the moving speed;

[0064] According to the distance from the current position of the moving target to the edge of the second area, which is known from the current time, and the moving speed, which is also known, the first time can be calculated. The first time is also the time when the moving target is expected to arrive at the second area;

[0065] Step S15: determining a starting period according to the first time, and monitoring the activity of the moving target in the interval monitoring area in the starting period, the starting period being a period before the first time and adjacent to the first time. For example, if the first time is 12:10, the starting period should be from 12:10 and last for a certain time. The time is generally determined according to the ratio of the interval road segment length between the first area and the second area and the moving speed;

[0066] Since the second region is distributed along the speed direction and located behind the speed of the first region, in the starting period immediately adjacent to the first time and before, the activity of the interval monitoring region where the starting detection guard target is located is monitored, which can exactly identify whether the active target stays in the first region or enters the second region. If it does not enter the second region, if it stays in the first region, it means that its action slows down obviously, and there may be a possibility of "crime" based on parked vehicles. If it has entered the second region before the end of the starting period, it indicates that it has theoretically accelerated its pace, so at this time, attention should be paid to the second region, so that the first region does not need to be monitored. Only in the middle position of the first region and the second region, it may be normal; it should be noted that the above determines that the active target value only comes from the intersection.

[0067] In this application, in the face of continuous guard targets (including parked vehicles and roadside portals), through the determination of the starting period, strictly according to the speed of the moving target, when monitoring the guard targets in combination with the first region and the second region, the activity of the active target can be determined without monitoring the active target all the time, (may or) without monitoring the guard target all the time, so as to determine whether the active target forms a potential threat to the guard target, thereby reducing the power consumption required for monitoring.

[0068] As shown in Figure 2 As a preferred embodiment of the present application, the target information of the active target is extracted according to the first image, which includes:

[0069] Step S111: identifying the active target in the image of the intersection position; the determination of the active target can use the foregoing target detection algorithm; common target detection algorithms include R-CNN, SPP-Net, Fast R-CNN, YOLOv1 and YOLOv2, etc., which can be combined with one or more in actual use; the active target generally refers to a person;

[0070] Step S112: frame processing is performed on the image, and the trend of the active target is determined according to at least two images; when frame processing, the frame rate can be tried until the different positions of the active target at consecutive time points are distinguished; taking at least two consecutive time points of different position changes can determine the trend;

[0071] Step S113: when the trend is towards the first road segment, the first feature of the active target is extracted;

[0072] When the trend is towards the first road segment, i.e. the moving target will enter the first road segment, the first feature can be extracted at this time; the image is subjected to feature extraction, and the first feature can be obtained, which can correspond to at least one of clothing, wearing and carrying objects; the feature extraction can utilize the following common image feature extraction algorithms, including HOG feature algorithm, LBP feature algorithm and Haar feature algorithm;

[0073] Step S114: judging whether the first feature meets the feature condition; the feature condition generally refers to whether the feature recognized at the intersection position is obvious enough, i.e. judging whether the first feature falls into the common feature library; for example, the clothing can be distinguished as having a hat (but there are several people with hats on this day, which has fallen into the common feature library), but the color cannot be distinguished;

[0074] Step S115: if yes, the first feature is determined as the target feature. When the feature recognized at the intersection position is obvious enough, the first feature falls into the common feature library, the feature condition is met, the first feature is determined as the target feature, and the moving target can be clearly and explicitly identified according to the first feature at this time;

[0075] As another case, as shown in Figure 3 the method further comprises:

[0076] Step S20: when the first feature does not meet the feature condition, a plurality of second road segments connected to the intersection position are identified;

[0077] Step S21: extracting a second image containing the first feature in the plurality of second road segments;

[0078] From the plurality of second road segments to the intersection position, the moving target can pass through one of the road segments, so by extracting the images of the plurality of second road segments (images before the time corresponding to the first image), and then by the first feature, the second image containing the first feature is preliminarily distinguished;

[0079] Step S22: screening a second sub-image in the second image that meets the clear condition; the further screening condition is the clear condition, and the clear condition includes a preset clarity, so that more or clearer features can be determined based on the second sub-image;

[0080] Step S23: identifying a second feature of the moving target in the second sub-image, and taking the combination of the first feature and the second feature as the target feature of the moving target.

[0081] The second feature is a complementary feature or another feature of the first feature, the complementary feature, for example, is a clothing color, such as a logo on the clothing; the another feature, such as a color of pants matched with the clothing, and the like; examples are not listed here; in fact, when the plurality of features are not obvious, a face contour or a facial portrait can be directly obtained, of course, this needs to obtain certain permission as a prerequisite, and thus the first feature and the second feature can be the face contour and the facial portrait, respectively;

[0082] The above can obtain more obvious target features by supplementing the second feature under necessary conditions, which is beneficial to quickly identify the moving target, and of course, can improve the accuracy of the moving target.

[0083] As shown in Figure 4 As a preferred embodiment of the present application, the monitoring of the moving target in the interval monitoring area in the starting period includes:

[0084] Step S151: issuing a starting instruction to the monitoring device of the roadside portal before the starting period to start the monitoring device and feed back an initial monitoring angle;

[0085] The initial monitoring angle is generally a monitoring angle covering a first area of the roadside portal;

[0086] Step S152: calculating a deflection angle of the monitoring device covering the parked vehicle according to the positional relationship between the roadside portal and the parked vehicle;

[0087] The positional relationship is obtained by displaying the coordinates of the parked vehicle on an electronic map, then determining the coordinate position of the parked vehicle relative to the roadside portal, and determining the first area of the parked vehicle on the electronic map according to the size ratio of the parked vehicle; the monitoring device generally exists in a position close to the roadside of the roadside portal, so the first and second areas can be regarded as linear areas, and a comparison of the size of the second area with the interval monitoring area excluding the second area in a certain direction (such as along the length direction of the first road section) is made to obtain a first ratio, the ratio of the initial monitoring angle to the deflection angle is greater than or equal to the first ratio; at least the first ratio is taken; the above can be used to calculate the deflection angle; for example, the first ratio is determined to be 0.25, and the initial monitoring angle is 50 degrees, then the deflection angle is 200 degrees, and the total coverage angle is 250 degrees.

[0088] Step S153: calculating a total coverage angle according to the initial monitoring angle and the deflection angle, and generating a monitoring instruction in the starting period according to the total coverage angle and the target feature, so that the monitoring device monitors the moving target according to the target feature.

[0089] The total coverage angle is generally the sum of the initial monitoring angle and the deflection angle. This means that the monitoring equipment needs to cover not only the original initial monitoring angle but also the deflection angle. Therefore, when there is only one monitoring equipment, the deflection angle of the monitoring equipment needs to be increased. For example, the monitoring equipment can be controlled to rotate at a first speed to rotate the initial monitoring angle and then rotate at a second speed to rotate the deflection angle. The first speed and the second speed can be equal. Increasing the monitoring height of the monitoring equipment is also possible. Alternatively, at least two monitoring equipment can be activated so that at least one other monitoring equipment can supplement the monitoring of the deflection angle (its monitoring angle is the deflection angle). However, it is important to note that at this time, at least two monitoring equipment should be controlled to monitor synchronously and report the monitoring results to ensure that the monitoring effect is equivalent to one monitoring equipment simultaneously covering the interval monitoring area.

[0090] Furthermore, when an active target is detected to meet preset conditions based on the target characteristics during the startup period, an abnormality alert is sent to the first terminal bound to the parked vehicle and the second terminal bound to the roadside portal. The preset conditions include that the active target is not located in the middle area between the first area and the second area.

[0091] It is important to understand that when monitoring active targets based on their characteristics, if an active target is detected, its current location is determined. When its location is outside the middle area, it indicates that the active target is either in the first area or the second area (e.g., entering a gate). It may initially have a "dangerous motive" towards parked vehicles or roadside gates, causing property damage, such as oil theft or other illegal activities. In addition, preset conditions may also include the active target being in the first or second area, and the active target carrying more items, or coming into contact with parked vehicles or roadside gates, thereby further enhancing the reliability of the abnormality alert.

[0092] In a preferred embodiment of the present invention, the method further includes:

[0093] Step: If no operation response is detected from either the first terminal or the second terminal within a set time period, an alarm message is issued through the monitoring device. The preset conditions include that the active target is not located in the middle area between the first area and the second area.

[0094] Understandably, when an abnormality alert is issued, provided that the preset conditions include the target not being in the middle of the first and second zones, the system begins detecting whether the user responds on both the first and second terminals. If a response is received, it indicates that the user is aware of the situation. The user is typically the vehicle owner or the owner of the access gate, or the person responsible for protection on the first road segment. If no response is received from either terminal, the user may not be aware of the information in a timely manner and may not be able to take timely protective patrol measures. In this case, a warning message is directly issued through the monitoring equipment, provided that the preset conditions include the target not being in the middle of the first and second zones. This could be a voice warning, such as "Please keep away from parked vehicles and roadside access gates, and do not loiter unnecessarily," to alert unrelated personnel and to alert the user.

[0095] like Figure 5 As shown, in another preferred embodiment of the present invention, an intelligent protection sensing system is provided, the system comprising:

[0096] The acquisition module 100 is used to acquire a first image, which includes an image of the intersection location;

[0097] Extraction module 200 is used to extract target information of a moving target based on the first image, the target information including movement speed and target features;

[0098] The identification module 300 is used to identify protected targets within the first road segment corresponding to the intersection location, the protected targets including parked vehicles and roadside gates;

[0099] The detection area determination module 400 is used to determine the interval monitoring area where the protected target is located. The interval monitoring area includes at least a first area and a second area distributed along the direction of the moving speed.

[0100] The estimation module 500 is used to estimate the first time when the moving target enters the second area based on the moving speed;

[0101] The timed start monitoring 600 is used to determine the start time period based on the first time, and to monitor the activity of the target in the interval monitoring area during the start time period. The start time period is the period before the first time and immediately adjacent to the first time.

[0102] like Figure 6 As shown, the extraction module 200 specifically includes:

[0103] The target recognition unit 2001 identifies moving targets in the image at the intersection location;

[0104] The framing and determination unit 2002 is used to perform framing processing on the image and determine the movement of the moving target based on at least two frames of images.

[0105] Extraction unit 2003 is used to extract the first feature of the active target when the trend is toward the first road segment;

[0106] Judgment unit 2004 is used to determine whether the first feature meets the feature conditions;

[0107] The condition determination unit 2005 is used to determine the first feature as the target feature if the first feature meets the feature conditions.

[0108] It should be noted that, referring to the specific implementation description of an intelligent protection sensing method in the foregoing embodiments, the implementation method of this system is completely consistent with that method, and will not be described again here.

[0109] The above embodiments of the present invention provide an intelligent protection sensing method, and based on the intelligent protection sensing method, an intelligent protection sensing system is provided. When facing continuous protection targets (including parked vehicles and roadside gates), by determining the start time period and strictly relying on the speed of the moving target, when monitoring protection targets in the first area and the second area, the activity status of the moving target can be determined without continuously monitoring the moving target, thereby reducing the power consumption required for monitoring.

[0110] In order for the above methods and systems to operate smoothly, the system may include more or fewer components than those described above, or combine certain components, or different components, in addition to the various modules mentioned above. For example, it may include input / output devices, network access devices, buses, processors, and memory.

[0111] The processor referred to can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the system, connecting various parts via various interfaces and lines.

[0112] The aforementioned memory can be used to store computer and system programs and / or modules. The aforementioned processor implements the various functions mentioned above by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as information collection template display function, product information publishing function, etc.). The data storage area may store data created based on the use of the berth status display system (such as product information collection templates corresponding to different product types, product information that different product providers need to publish, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0113] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0114] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0115] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0116] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent protection awareness method, characterized in that, The method comprises: acquiring a first image, the first image comprising an image of a position of an intersection; extracting target information of a moving target according to the first image, the target information comprising a moving speed and a target feature; identifying a protection target in a first section corresponding to the position of the intersection, the protection target comprising a parked vehicle on a roadside and a roadside portal; determining an interval monitoring area in which the protection target is located, the interval monitoring area comprising at least a first area and a second area distributed along a direction of the moving speed; estimating a first time at which the moving target enters the second area according to the moving speed; determining a starting period according to the first time, and monitoring an activity of the moving target in the interval monitoring area in the starting period, the starting period being a period before and immediately adjacent to the first time; the extracting of the target information of the moving target according to the first image comprises: identifying the moving target in the image of the position of the intersection; performing frame processing on the image to determine a trend of the moving target according to at least two frames of images; extracting a first feature of the moving target when the trend is towards the first section; judging whether the first feature meets a feature condition; if yes, determining the first feature as the target feature; the method further comprises: when the first feature does not meet the feature condition, identifying a plurality of second sections connected to the position of the intersection; extracting a second image containing the first feature in the plurality of second sections; screening a second sub-image meeting a clear condition in the second image; identifying a second feature of the moving target in the second sub-image, and combining the first feature and the second feature as the target feature of the moving target; the monitoring of the activity of the moving target in the interval monitoring area in the starting period comprises: issuing a starting instruction to a monitoring device of the roadside portal before the starting period to start the monitoring device and feed back an initial monitoring angle; calculating a deflection angle of the monitoring device covering the parked vehicle according to a positional relationship between the roadside portal and the parked vehicle; calculating a total coverage angle according to the initial monitoring angle and the deflection angle, and generating a monitoring instruction in the starting period according to the total coverage angle and the target feature, so that the monitoring device monitors the moving target according to the target feature. 2.The intelligent guard perception method of claim 1, wherein, the method further comprises: when detecting that the moving target meets a preset condition according to the target feature in the starting period, sending an abnormal prompt information to a first terminal bound to the parked vehicle and a second terminal bound to the roadside portal, the preset condition comprising that the moving target is not in a middle area between the first area and the second area. 3.The intelligent guard awareness method of claim 2, wherein, the method further comprises: when no operation response of any one of the first terminal and the second terminal is detected within a set time length, issuing an alarm information through the monitoring device, the preset condition comprising that the moving target is not in a middle area between the first area and the second area.

4. An intelligent protective perception system, characterized in that, The system is used to implement the method of claim 1, comprising: an acquisition module, configured to acquire a first image, the first image comprising an image of a position of an intersection; an extraction module, configured to extract target information of a moving target according to the first image, the target information comprising a moving speed and a target feature; The identification module is configured to identify a guard object in a first road section corresponding to the intersection position, the guard object including a parked vehicle on a roadside and a roadside portal; The detection region determination module is configured to determine an interval monitoring region in which the guard object is located, the interval monitoring region including at least a first region and a second region distributed along a direction of the moving speed; The estimation module is configured to estimate a first time at which the moving object enters the second region according to the moving speed; The timing start monitoring is configured to determine a start period according to the first time, and monitor a moving state of the moving object in the interval monitoring region in the start period, the start period being a period before and adjacent to the first time.

5. The intelligent guard sensing system of claim 4, wherein, The extraction module specifically includes: A target identification unit configured to identify a moving object in an image of the intersection position; A frame division and determination unit configured to perform frame division processing on the image, and determine a moving trend of the moving object according to at least two frames of images; An extraction unit configured to extract a first feature of the moving object when the moving trend is toward the first road section; A judgment unit configured to judge whether the first feature meets a feature condition; A condition determination unit configured to determine the first feature as a target feature if the first feature meets the feature condition.

Images