Inductive control method, electronic equipment and computer readable storage medium

Abstract

The invention provides an inductive control method, which is applied to edge computing equipment and comprises the following steps: acquiring operation data of a traffic signal lamp of a target intersection at the current moment sent by a signal machine; the operation data comprises an operation scheme, an execution target stage and a lamp state of each phase in the target stage; acquiring video stream data at the current moment acquired by at least one video monitoring device arranged at the target intersection; according to the operation scheme and the video stream data, determining a control signal value of each phase of which the lamp state is a green lamp in the target stage; and sending a control instruction comprising the control signal value of each phase to the signal machine. In this way, the passing efficiency of vehicles and the safety of pedestrians crossing the street can be effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of traffic control, and in particular to an induction control method, an electronic device and a computer-readable storage medium. Background Art

[0002] With the improvement of people's living standards, cars have become an essential tool for people to travel. However, the continuous increase in urban cars has also brought huge loads to urban road traffic. Therefore, it is necessary to control urban traffic lights to alleviate urban traffic congestion and avoid frequent traffic accidents.

[0003] In the related art, the traffic lights at the intersections are usually displayed according to the configured cycle duration and duration. However, when the traffic situation changes suddenly, the traffic lights cannot respond quickly to the changes in urban traffic, and the control of the traffic lights is not flexible enough, which affects the traffic efficiency of vehicles and the safety of pedestrians crossing the street. Summary of the invention

[0004] In response to the above technical problems, the present application provides a sensing control method, an electronic device and a computer-readable storage medium, which can effectively improve the traffic efficiency of vehicles and the safety of pedestrians crossing the street.

[0005] The present application provides an induction control method, which is applied to an edge computing device, including:

[0006] Obtaining the operation data of the traffic lights at the target intersection sent by the signal machine at the current moment; the operation data includes the operation plan, the target stage of execution and the light state of each phase in the target stage;

[0007] Acquire video stream data at a current moment collected by at least one video surveillance device disposed at a target intersection;

[0008] According to the operation plan and the video stream data, determine the control signal value of each phase of the light state being green in the target phase;

[0009] A control instruction including a control signal value for each phase is sent to the signal.

[0010] In one embodiment, the operation scheme includes computing control logic and target object quantity thresholds corresponding to each phase in the target phase;

[0011] According to the operation plan and the video stream data, the control signal value of each phase of the light state of the target stage is determined, including:

[0012] Performing phase correlation detection on the video stream data to obtain detection results of the video stream sub-data associated with each phase in which the light state is green in the target phase;

[0013] According to the operation control logic, the target object quantity threshold corresponding to each phase and the detection result of the video stream sub-data associated with each phase, the control signal value of each phase whose light state is green in the target stage is determined.

[0014] In one embodiment, phase correlation detection is performed on the video stream data to obtain detection results of the video stream sub-data associated with each phase in which the light state is green in the target phase, including:

[0015] Determine the target video stream sub-data associated with the target phase from the video stream data; the target phase is any phase in the target phase where the light state is green;

[0016] The target objects in the target video stream sub-data that are in the target detection area corresponding to the target phase are identified and detected to obtain the number of target objects in the target detection area corresponding to the target phase.

[0017] In one embodiment, the control signal value of each phase in which the light state is green in the target phase is determined according to the operation control logic, the target object quantity threshold corresponding to each phase, and the detection result of the video stream sub-data associated with each phase, including:

[0018] According to the operation control logic and the target object quantity threshold corresponding to each phase in the target phase and the detection results of the video stream sub-data associated with each phase, the offline state value and the overflow state value of each phase with a green light in the target phase and the empty state value of the target phase are determined;

[0019] According to the minimum decision time and maximum decision time of the target phase, the offline state value and overflow state value of each phase with the light state of green light in the target phase, and the idle state value of the target phase, determine the offline signal value and overflow signal value of each phase with the light state of green light in the target phase, as well as the idle signal value and delayed signal value of the target phase at the current moment; the minimum decision time and the maximum decision time are determined based on the minimum green light duration and the maximum green light duration of all phases in the target phase;

[0020] According to the offline signal value and overflow signal value of each phase with the light state of green light in the target stage and the idle signal value and delayed signal value of the target stage at the current moment, the control signal value of each phase with the light state of green light in the target stage is determined.

[0021] In one embodiment, the target objects include vehicles and pedestrians; the detection area includes an import vehicle detection area, an export vehicle detection area, and a pedestrian detection area; the target object quantity threshold includes an import vehicle quantity threshold, an export vehicle quantity threshold, and a pedestrian quantity threshold; the empty state value includes an import state value and a sidewalk state value;

[0022] According to the operation control logic and the target object quantity threshold corresponding to each phase in the target stage and the detection result of the video stream sub-data associated with each phase, the offline state value and the overflow state value of each phase with the light state of green light in the target stage and the empty state value of the target stage are determined, including:

[0023] If the detection result indicates that the video stream sub-data associated with the target phase is not obtained, then the offline state value of the target phase is marked as a first preset value, otherwise the offline state value of the target phase is marked as a second preset value;

[0024] If the detection result indicates that the number of vehicles in the exit vehicle detection zone corresponding to the target phase is greater than the exit vehicle number threshold, the overflow state value of the target phase is marked as a first preset value, otherwise the overflow state value of the target phase is marked as a second preset value;

[0025] When the operation control logic is the first operation logic, if the detection result indicates that the number of vehicles in the import vehicle detection area corresponding to each phase in which the light state is green in the target stage is respectively greater than the corresponding import vehicle number threshold, then the import state value of the target stage is marked as the first preset value, otherwise the import state value of the target stage is marked as the second preset value; if the detection result indicates that the number of pedestrians in the pedestrian detection area corresponding to each phase in which the light state is green in the target stage is respectively greater than the corresponding pedestrian number threshold, then the sidewalk state value of the target stage is marked as the first preset value, otherwise the sidewalk state value of the target stage is marked as the second preset value;

[0026] When the operation control logic is the second operation logic, if the detection result indicates that the sum of the number of vehicles in the import vehicle detection area corresponding to each phase in which the light state is green in the target stage is greater than the sum of the corresponding import vehicle number thresholds, then the import state value of the target stage is marked as the first preset value, otherwise the import state value of the target stage is marked as the second preset value; if the detection result indicates that the sum of the number of pedestrians in the pedestrian detection area corresponding to each phase in which the light state is green in the target stage is greater than the sum of the corresponding pedestrian number thresholds, then the sidewalk state value of the target stage is marked as the first preset value, otherwise the sidewalk state value of the target stage is marked as the second preset value.

[0027] In one embodiment, according to the minimum decision time, the maximum decision time, the offline state value and the overflow state value of each phase with the light state of green light in the target stage, and the empty release state value of the target stage, the offline signal value and the overflow signal value of each phase with the light state of green light in the target stage and the empty release signal value and the delayed signal value of the target stage at the current moment are determined, including:

[0028] If the minimum decision time of the target stage is not reached at the current moment, the offline signal value and the overflow signal value of each phase whose light state is green in the target stage are determined to be the second preset value and the empty release signal value and the delay signal value of the target stage at the current moment are determined to be the first preset value and the second preset value;

[0029] If the current moment has reached the minimum decision moment of the target stage but has not reached the maximum decision moment of the target stage, then when the number of consecutive times that the offline state value of the target phase is the first preset value reaches the preset maximum offline number, the offline signal value of the target phase is marked as the third preset value, otherwise the offline signal value of the target phase is marked as the second preset value; when the number of consecutive times that the overflow state value of the target phase is the first preset value reaches the preset maximum overflow number, the overflow signal value of the target phase is marked as the fourth preset value, otherwise the overflow signal value of the target phase is marked as the second preset value; when the number of consecutive times that the import state value of the target stage is the second preset value reaches the preset maximum empty release number and the sidewalk state value is the second preset value, the empty release signal value of the target stage at the current moment is marked as the second preset value, otherwise the empty release signal value of the target stage at the current moment is marked as the first preset value; when the difference between the current moment and the start moment when the empty release signal value of the target stage is the second preset value is less than or equal to the preset delay time, the delay signal value of the target stage at the current moment is marked as the first preset value, otherwise the delay signal value of the target stage at the current moment is marked as the second preset value;

[0030] If the maximum decision moment of the target stage has been reached at the current moment, the offline signal value and the overflow signal value of each phase with the light state of green light in the target stage are determined to be the second preset value and the second preset value, and the idle signal value and the delayed signal value of the target stage at the current moment are determined to be the second preset value and the second preset value.

[0031] In one embodiment, according to the offline signal value and overflow signal value of each phase whose light state is green in the target stage and the empty release signal value and delayed signal value of the target stage at the current moment, the control signal value of each phase whose light state is green in the target stage is determined, including:

[0032] Detecting whether the control signal value of the target phase at the previous moment is a preset signal value;

[0033] If yes, the preset signal value is used as the control signal value of the target phase at the current moment;

[0034] If not, the maximum value among the offline signal value, overflow signal value of the target phase at the current moment, the empty signal value and the delayed signal value of the target phase at the current moment is used as the control signal value of the target phase at the current moment.

[0035] The present application also provides an induction control method, which is applied to a signal machine, comprising:

[0036] Send the current operation data of the traffic lights at the target intersection to the edge computing device; the operation data includes the operation plan, the target stage of execution, and the light state of each phase in the target stage;

[0037] Obtaining a control instruction sent by the edge computing device including a control signal value of each phase in which the light state is green in the target phase;

[0038] Control the traffic lights at the target intersection to execute control instructions.

[0039] The present application also provides an electronic device, comprising: a memory and a processor, wherein the memory stores computer program instructions for execution on the processor, and when the processor executes the computer program instructions, the induction control method as described above is implemented.

[0040] The present application also provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the induction control method as described above is implemented.

[0041] As described above, in the induction control method provided by the present application, by acquiring the video stream data collected by the existing video surveillance equipment at the target intersection, and combining the operating data at the current moment, the control signal value of each phase when the light state is green in the target stage is determined, so that the traffic light can control the traffic flow at the target intersection according to the control signal value of each phase, which can effectively improve the traffic efficiency of vehicles and the safety of pedestrians crossing the street, and has flexible control and low application cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a phase diagram;

[0043] Figure 2 is a schematic diagram of the detection area;

[0044] Figure 3 A schematic diagram of a flow chart of an induction control method provided in an embodiment of the present application;

[0045] Figure 4 A schematic flow chart of another induction control method provided in an embodiment of the present application;

[0046] Figure 5 A schematic diagram of the architecture of a sensing control system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] First, the relevant terms involved in the embodiments of the present application are explained.

[0049] Phase: used to indicate the direction and turn information of at least one traffic flow that obtains the right of way at the same time. Each phase corresponds to a unique phase identifier, which can be a phase number. For example, a number represents the phase identifier. Figure 1 As shown, phase 1 indicates a left turn from the east, phase 2 indicates going straight from the east, phase 3 indicates a right turn from the east, phase 4 indicates pedestrians on the east side, phase 5 indicates a left turn from the west, phase 6 indicates going straight from the west, phase 7 indicates a right turn from the west, phase 8 indicates pedestrians on the west side, phase 9 indicates a left turn from the south, phase 10 indicates going straight from the south, phase 11 indicates a right turn from the south, phase 12 indicates pedestrians on the south side, phase 13 indicates a left turn from the north, phase 14 indicates going straight from the north, phase 15 indicates a right turn from the north, and phase 16 indicates pedestrians on the north side.

[0050] Phase: The naming of phase combinations, including east-west straight, east-west left turn, north-south straight, north-south left turn, east phase release, west phase release, south phase release, north phase release, etc. Taking the east-west straight phase as an example, it includes the east straight phase and the west straight phase.

[0051] Traffic plan: A combination of phase sequence and phase duration. For example, for a traffic plan, the phases included are north-south straight, north phase release, north-south straight, east phase release, and the phase duration of each phase is 30 seconds.

[0052] Detection area: The area that needs to be detected in the phase of the intersection, including the entry vehicle detection area, exit vehicle detection area and pedestrian detection area. The entry vehicle detection area is used to detect vehicles about to enter the intersection, the exit vehicle detection area is used to detect vehicles that have entered the intersection but have not yet exited the intersection (i.e., between the stop lines on both sides of the intersection), and the pedestrian detection area is used to detect pedestrians on the sidewalk. Take the phase of the east-west straight line as an example, Figure 2 As shown, the east-west straight driving stage includes the east straight driving phase and the west straight driving phase. The detection areas corresponding to the east straight driving phase include the import vehicle detection area 1, the exit vehicle detection area 1 and the pedestrian detection area 1. The detection areas corresponding to the west straight driving phase include the import vehicle detection area 2, the exit vehicle detection area 2 and the pedestrian detection area 2.

[0053] This embodiment provides an induction control method, such as Figure 3 As shown, the induction control method can be applied to edge computing devices, including:

[0054] Step S101, obtaining the operation data of the traffic lights at the target intersection at the current moment; the operation data includes the operation plan, the target stage of execution and the light state of each phase in the target stage.

[0055] Step S102: Obtain video stream data at the current moment collected by at least one video surveillance device installed at the target intersection.

[0056] Step S103: Determine the control signal value of each phase of the light state of the green light in the target phase according to the operation plan and the video stream data.

[0057] Step S104: Send a control instruction including the control signal value of each phase to the signal machine.

[0058] The target intersection is an intersection that needs to be controlled by induction, which can be a cross intersection, a T-shaped intersection or other forms of intersections, which are not specifically limited here. Video surveillance equipment includes but is not limited to electric police, checkpoints, traffic cameras and radar vision equipment to provide stable video stream data for edge computing devices. The edge computing device can obtain video stream data directly from the video surveillance device, or obtain video stream data collected by the video surveillance device from a video terminal box. At different time periods, the traffic plan of the traffic lights at the target intersection may be different. For example, the traffic plan corresponding to 7 am to 9 am may be the morning peak traffic plan, while the traffic plan corresponding to 9 am to 12 am may be the flat peak traffic plan, etc. At the same time, the target phases executed at different times may also be different. For example, at 8:10:10 am, the target phase to be executed may be north-south straight, while at 8:10:40 am, the target phase to be executed may be north-south left turn, etc. The phase in the target phase is a phase related to the target phase. For example, when the target phase is north-south straight driving, the phase in the target phase includes north straight driving phase, south straight driving phase, east pedestrian phase, and west pedestrian phase.

[0059] The operation scheme is used to indicate the traffic control rules for traffic flows in each direction, and may include the traffic scheme, operation control logic, and the target object quantity threshold, minimum green light duration, maximum green light duration, and delayed output duration corresponding to each phase in the target stage. When the edge computing device has stored different operation schemes for the traffic lights at the target intersection, the operation scheme in the operation data may be an operation scheme identifier such as an operation scheme number, and the edge computing device may determine the corresponding operation scheme locally based on the operation scheme identifier.

[0060] Among them, according to the operation plan and video stream data of the traffic lights at the target intersection at the current moment, a control signal value for the phase in which the light is green in the target stage can be generated, so that the state of the target stage or phase can be adjusted based on the control signal value through the traffic light, thereby improving the safety of pedestrians crossing the street and the efficiency of vehicle traffic.

[0061] In one embodiment, the operation scheme includes computing control logic and target object quantity thresholds corresponding to each phase in the target stage; determining the control signal value of each phase in the target stage where the light state is green according to the operation scheme and the video stream data, including:

[0062] Performing phase correlation detection on the video stream data to obtain detection results of the video stream sub-data associated with each phase in which the light state is green in the target phase;

[0063] According to the operation control logic, the target object quantity threshold corresponding to each phase and the detection result of the video stream sub-data associated with each phase, the control signal value of each phase whose light state is green in the target stage is determined.

[0064] Among them, the operation control logic is used to indicate the operation logic in the process of generating the control signal value of the phase, including the first operation logic and the second operation logic. The target objects include vehicles and pedestrians, and different phases are used to control vehicles or pedestrians respectively, and the target object quantity thresholds corresponding to different phases may be the same or may be different. It can be understood that since the video stream sub-data related to the control of different phases are different, for example, the video stream sub-data related to the control of the north straight phase needs to include the video data of the lanes, intersections and sidewalks in the north-south straight direction, etc., therefore, it is necessary to perform phase correlation detection on the video stream data to obtain the detection results of the video stream sub-data associated with each phase in the target stage where the light state is green. The detection result may include whether the phase-associated video stream sub-data is obtained, and when the phase-associated video stream sub-data is obtained, the number of target objects in the detection area corresponding to the phase, etc.

[0065] In one embodiment, phase correlation detection is performed on the video stream data to obtain detection results of the video stream sub-data associated with each phase in which the light state is green in the target phase, including:

[0066] Determine the target video stream sub-data associated with the target phase from the video stream data; the target phase is any phase in the target phase where the light state is green;

[0067] The target objects in the target video stream sub-data that are in the target detection area corresponding to the target phase are identified and detected to obtain the number of target objects in the target detection area corresponding to the target phase.

[0068] It can be understood that although the shooting ranges of different video surveillance devices installed at the target intersection are different, it can be determined which one or several video surveillance devices can be used to shoot and obtain the video stream sub-data associated with each phase according to the shooting ranges of different video surveillance devices. Figure 2For the east straight phase, the video stream data captured by the video surveillance equipment set on the east side of the target intersection and whose shooting range covers the import vehicle detection area 1, the exit vehicle detection area 1 and the pedestrian detection area 1 can be used as the associated video stream sub-data. After determining the target video stream sub-data associated with the target phase, the vehicles in the vehicle detection area corresponding to the target phase can be identified and detected based on the vehicle recognition algorithm to obtain the number of vehicles in the vehicle detection area corresponding to the target phase. The pedestrians in the pedestrian detection area corresponding to the target phase can also be identified and detected based on the pedestrian recognition algorithm to obtain the number of pedestrians in the pedestrian detection area corresponding to the target phase. It should be noted that before identifying and detecting the target object in the target detection area corresponding to the target phase in the target video stream sub-data, a rectangle can be set as the analysis range of the video stream screen based on the detection requirements to minimize interference. At the same time, when setting the exit vehicle detection area, in order to make the exit vehicle detection area cover the vehicles that have entered the intersection as much as possible, the shape of the exit vehicle detection area can be set to a polygon, such as a hexagon or an octagon.

[0069] In one embodiment, the control signal value of each phase in which the light state is green in the target phase is determined according to the operation control logic, the target object quantity threshold corresponding to each phase, and the detection result of the video stream sub-data associated with each phase, including:

[0070] According to the operation control logic and the target object quantity threshold corresponding to each phase in the target phase and the detection results of the video stream sub-data associated with each phase, the offline state value and the overflow state value of each phase with the light state of green light in the target phase and the empty state value of the target phase are determined; the empty state value includes the import state value and the sidewalk state value;

[0071] According to the minimum decision time and maximum decision time of the target stage, the offline state value and overflow state value of each phase with the light state of green light in the target stage, and the idle state value of the target stage, determine the offline signal value and overflow signal value of each phase with the light state of green light in the target stage, as well as the idle signal value and delayed signal value of the target stage at the current moment; the minimum decision time and the maximum decision time are determined based on the minimum green light duration, the maximum green light duration and the delayed output duration corresponding to all phases in the target stage respectively;

[0072] According to the offline signal value and overflow signal value of each phase with the light state of green light in the target stage and the idle signal value and delayed signal value of the target stage at the current moment, the control signal value of each phase with the light state of green light in the target stage is determined.

[0073] Among them, for any phase in the target stage where the light state is green, the offline state value and overflow state value of the phase can be determined according to the operation control logic, the target object quantity threshold corresponding to the phase in the target stage, and the detection result of the video stream sub-data associated with the phase. The offline state value of the phase is used to characterize whether the video stream sub-data associated with the phase is obtained, that is, whether there is a video surveillance device offline and the corresponding video stream sub-data cannot be obtained. The overflow state value of the phase is used to characterize the size of the number of vehicles in the exit vehicle detection area corresponding to the phase. In addition, according to the target object quantity threshold corresponding to each phase in the target stage and the detection result of the video stream sub-data associated with each phase, combined with the operation control logic, the empty state value of the target stage at the current moment can be determined. The empty state value includes the import state value and the sidewalk state value. The import state value of the target stage is used to characterize the size of the number of vehicles in the import vehicle detection area corresponding to all phases in the target stage, and the sidewalk state value of the target stage is used to characterize the size of the number of pedestrians in the pedestrian detection area corresponding to all phases in the target stage. The minimum decision time and the maximum decision time of the target stage are used to indicate the time range in which the target stage can be adjusted and controlled.

[0074] Among them, the operation plan includes the minimum green light duration, maximum green light duration and delayed output duration corresponding to all phases in the target stage. The minimum green light duration is used to indicate the minimum duration when the indicator state is green, and the maximum green light duration is used to indicate the maximum duration when the indicator state is green. At the first moment of entering the target stage (that is, the previous moment does not belong to another stage), the minimum decision moment of the target stage can be determined based on the minimum decision duration of the target stage, and the maximum decision moment of the target stage can be determined based on the maximum decision duration of the target stage. The minimum decision duration of the target stage is the value obtained by subtracting the maximum delayed output duration and the preset duration from the minimum value of the minimum green light duration of all phases in the target stage. The maximum decision duration of the target stage is the value obtained by subtracting the maximum delayed output duration and the preset duration from the minimum value of the maximum green light duration of all phases in the target stage. The preset duration can be set according to actual needs, such as 4 seconds. For example, assuming that the current time 9:10:10 is the first moment of the target stage, the minimum decision time of the target stage is 20 seconds, and the maximum decision time is 30 seconds, then the minimum decision time of the target stage is 9:10:30, and the maximum decision time of the target stage is 9:10:40.

[0075] In one embodiment, the target objects include vehicles and pedestrians; the detection area includes an import vehicle detection area, an export vehicle detection area, and a pedestrian detection area; the target object quantity threshold includes an import vehicle quantity threshold, an export vehicle quantity threshold, and a pedestrian quantity threshold; and the empty state value includes an import state value and a sidewalk state value. According to the operation control logic and the target object quantity threshold corresponding to each phase in the target stage and the detection results of the video stream sub-data associated with each phase, the offline state value and the overflow state value of each phase with a green light in the target stage and the empty state value of the target stage are determined, including:

[0076] If the detection result indicates that the video stream sub-data associated with the target phase is not obtained, then the offline state value of the target phase is marked as a first preset value, otherwise the offline state value of the target phase is marked as a second preset value;

[0077] If the detection result indicates that the number of vehicles in the exit vehicle detection zone corresponding to the target phase is greater than the exit vehicle number threshold, the overflow state value of the target phase is marked as a first preset value, otherwise the overflow state value of the target phase is marked as a second preset value;

[0078] When the operation control logic is the first operation logic, if the detection result indicates that the number of vehicles in the import vehicle detection area corresponding to each phase in which the light state is green in the target stage is respectively greater than the corresponding import vehicle number threshold, then the import state value of the target stage is marked as the first preset value, otherwise the import state value of the target stage is marked as the second preset value; if the detection result indicates that the number of pedestrians in the pedestrian detection area corresponding to each phase in which the light state is green in the target stage is respectively greater than the corresponding pedestrian number threshold, then the sidewalk state value of the target stage is marked as the first preset value, otherwise the sidewalk state value of the target stage is marked as the second preset value;

[0079] When the operation control logic is the second operation logic, if the detection result indicates that the sum of the number of vehicles in the import vehicle detection area corresponding to each phase in which the light state is green in the target stage is greater than the sum of the corresponding import vehicle number thresholds, then the import state value of the target stage is marked as the first preset value, otherwise the import state value of the target stage is marked as the second preset value; if the detection result indicates that the sum of the number of pedestrians in the pedestrian detection area corresponding to each phase in which the light state is green in the target stage is greater than the sum of the corresponding pedestrian number thresholds, then the sidewalk state value of the target stage is marked as the first preset value, otherwise the sidewalk state value of the target stage is marked as the second preset value.

[0080] The first preset value can be set to 1, and the second preset value can be set to 0. The first operation logic can be an AND logic, that is, each case needs to satisfy the corresponding conditions separately; the second operation logic can be an ADD logic, that is, all cases need to satisfy the sum of the corresponding conditions. Figure 2, taking the target stage as the east-west straight driving as an example, the east-west straight driving stage includes the east straight driving phase and the west straight driving phase. The detection areas corresponding to the east straight driving phase include the import vehicle detection area 1, the exit vehicle detection area 1 and the pedestrian detection area 1, and the detection areas corresponding to the west straight driving phase include the import vehicle detection area 2, the exit vehicle detection area 2 and the pedestrian detection area 2. When the operation control logic is the first operation logic and the first operation logic is the sum logic, if the number of vehicles in the import vehicle detection area 1 is greater than the first import vehicle number threshold, and the number of vehicles in the import vehicle detection area 2 is greater than the second import vehicle number threshold, then the import state value of the target stage is marked as 1, otherwise the import state value of the target stage is marked as 0. At the same time, if the number of pedestrians in the pedestrian detection area 1 is greater than the first pedestrian number threshold and the pedestrian detection area 2 is greater than the second pedestrian number threshold, then the import state value of the target stage is marked as 0. If the number of pedestrians in the imported vehicle detection area one is greater than the second pedestrian number threshold, the sidewalk state value of the marked target stage is 1, otherwise the sidewalk state value of the marked target stage is 0; when the operation control logic is the second operation logic and the second operation logic is the addition logic, if the sum of the number of vehicles in the imported vehicle detection area one and the number of vehicles in the imported vehicle detection area two is greater than the sum of the first imported vehicle number threshold and the second imported vehicle number threshold, the import state value of the marked target stage is 1, otherwise the import state value of the marked target stage is 0, and at the same time, if the sum of the number of pedestrians in the pedestrian detection area one and the number of pedestrians in the pedestrian detection area two is greater than the sum of the first pedestrian number threshold and the second pedestrian number threshold, the sidewalk state value of the marked target stage is 1, otherwise the sidewalk state value of the marked target stage is 0.

[0081] In one embodiment, according to the minimum decision time, the maximum decision time, the offline state value and the overflow state value of each phase with the light state of green light in the target stage, and the empty release state value of the target stage, the offline signal value and the overflow signal value of each phase with the light state of green light in the target stage and the empty release signal value and the delayed signal value of the target stage at the current moment are determined, including:

[0082] If the minimum decision time of the target stage is not reached at the current moment, the offline signal value and the overflow signal value of each phase whose light state is green in the target stage are determined to be the second preset value and the empty release signal value and the delay signal value of the target stage at the current moment are determined to be the first preset value and the second preset value;

[0083] If the current moment has reached the minimum decision moment of the target stage but has not reached the maximum decision moment of the target stage, then when the number of consecutive times that the offline state value of the target phase is the first preset value reaches the preset maximum offline number, the offline signal value of the target phase is marked as the third preset value, otherwise the offline signal value of the target phase is marked as the second preset value; when the number of consecutive times that the overflow state value of the target phase is the first preset value reaches the preset maximum overflow number, the overflow signal value of the target phase is marked as the fourth preset value, otherwise the overflow signal value of the target phase is marked as the second preset value; when the number of consecutive times that the import state value of the target stage is the second preset value reaches the preset maximum empty release number and the sidewalk state value is the second preset value, the empty release signal value of the target stage at the current moment is marked as the second preset value, otherwise the empty release signal value of the target stage at the current moment is marked as the first preset value; when the difference between the current moment and the start moment when the empty release signal value of the target stage is the second preset value is less than or equal to the preset delay time, the delay signal value of the target stage is marked as the first preset value, otherwise the delay signal value of the target stage is marked as the second preset value;

[0084] If the maximum decision moment of the target stage has been reached at the current moment, the offline signal value and the overflow signal value of each phase with the green light in the target stage are determined to be the second preset value and the second preset value, and the idle signal value and the delayed signal value of the target stage at the current moment are determined to be the second preset value.

[0085] For example, assuming that the minimum decision-making time of the target stage is 9:10:30 and the maximum decision-making time of the target stage is 9:10:40, if the current time is 9:10:20, it means that the current time has not reached the minimum decision-making time of the target stage; if the current time is 9:10:30, it means that the current time has reached the minimum decision-making time of the target stage but has not reached the maximum decision-making time of the target stage; if the current time is 9:10:40, it means that the current time has reached the maximum decision-making time of the target stage.

[0086] Among them, the third preset value can be set to 2, the fourth preset value can be set to 3, etc. The preset maximum number of offline times, the preset maximum number of overflow times, and the preset maximum number of empty release times can be set according to actual needs, and are not specifically limited here. It can be understood that in order to ensure that due to the influence of factors such as the distance of the detection area from the stop line of the lane and the interference of pedestrians and non-motor vehicles at different time periods, the target stage needs to be terminated for a delay period to ensure that the detected vehicle passes without stopping. The delay signal value of the target stage at the current moment can be determined based on the current moment, the empty release signal value of the target stage is the start time of the second preset value, the preset delay time, etc. Among them, if the empty release signal value of the target stage has been the first preset value before and at the current moment, the delay signal value of the target stage is the second preset value, then the delay signal value of the target stage is marked as the second preset value.

[0087] In one embodiment, according to the offline signal value and overflow signal value of each phase whose light state is green in the target stage and the empty release signal value and delayed signal value of the target stage at the current moment, the control signal value of each phase whose light state is green in the target stage is determined, including:

[0088] Detecting whether the control signal value of the target phase at the previous moment is a preset signal value;

[0089] If yes, the preset signal value is used as the control signal value of the target phase at the current moment;

[0090] If not, the maximum value among the offline signal value, overflow signal value of the target phase at the current moment, the empty signal value and the delayed signal value of the target phase at the current moment is used as the control signal value of the target phase at the current moment.

[0091] Among them, the preset signal value can be set according to actual needs, and can be a second preset value or a fourth preset value, etc. When the control signal value of the target phase at the previous moment is not the preset signal value, the maximum value of the offline signal value, overflow signal value of the target phase at the current moment, and the empty signal value and delayed signal value of the target stage at the current moment is used as the control signal value of the target phase at the current moment. For example, when the target stage only includes a single phase with a green light state, when the control signal value is the first preset value, it indicates a green light, that is, continuing the current stage; when the control signal value is the second preset value, it indicates turning off the light, that is, ending the current stage; when the control signal value is the third preset value, it indicates that the video surveillance equipment is offline and is handled by the signal machine itself; when the control signal value is the fourth preset value, it indicates an overflow, that is, closing the associated phase of the overflow. When the target stage includes two phases with a green light state, the control operation of the signal machine on the traffic light can be determined based on the control signal values ​​of the two phases. For example, continue to refer to Figure 2 Taking the target stage as the east-west straight travel as an example, the east-west straight travel stage includes the east straight travel phase and the west straight travel phase. When the control signal values ​​corresponding to the east straight travel phase and the west straight travel phase are both the first preset values, it means that the target stage may need to be stopped, and when the control signal values ​​corresponding to the east straight travel phase and the west straight travel phase are both the second preset values, it means that the target stage can be continued, etc.

[0092] In summary, in the induction control method provided in the above embodiment, by acquiring the video stream data collected by the existing video surveillance equipment at the target intersection, and combining the operating data at the current moment, the control signal value of each phase when the light state is green in the target stage is determined, so that the traffic light can control the traffic flow at the target intersection according to the control signal value of each phase, which can effectively improve the traffic efficiency of vehicles and the safety of pedestrians crossing the street, and has flexible control and low application cost.

[0093] Based on the same inventive concept as the above embodiments, the present application also provides another inductive control method, such as Figure 4 As shown, the induction control method can be applied to signal machines, including:

[0094] Step S201: Sending the current operation data of the traffic lights at the target intersection to the edge computing device; the operation data includes the operation plan, the target stage of execution, and the light state of each phase in the target stage;

[0095] Step S202: obtaining a control instruction sent by the edge computing device including a control signal value of each phase whose light state is green in the target phase;

[0096] Step S203: Control the traffic light at the target intersection to execute the control instruction.

[0097] Among them, the signal machine and the edge computing device are connected in communication, for example, the signal machine and the edge computing device can communicate through protocol messages. The signal machine may include a local fixed-cycle control mode and a sensing control mode. In the local fixed-cycle control mode, the signal machine will control the traffic lights to execute different traffic plans according to the preset cycle, while in the sensing control mode, the signal machine will control the traffic lights according to the control instructions sent by the edge computing device. The signal machine can send the current operation data to the edge computing device periodically or in real time; the operation data includes the operation plan, the target stage of execution, and the light state of each phase in the target stage.

[0098] Among them, after receiving the control instruction sent by the edge computing device including the control signal value of each phase with the light state of green light in the target stage, the traffic light at the corresponding control target intersection executes the control instruction. For example, when the target stage only includes a single phase with the light state of green light, when the control signal value of the phase is the first preset value, the traffic light continues to be controlled to maintain the current stage, that is, the light state of the control phase is kept as green light; when the control signal value of the phase is the second preset value, the traffic light is controlled to end the current stage, that is, the light state of the control phase is adjusted from green light to red light; when the control signal value of the phase is the third preset value, it indicates that the video surveillance device associated with the phase is offline, and the signal machine handles it by itself; when the control signal value of the phase is the fourth preset value, the traffic light is controlled to close the overflowed associated phase. Among them, when the video surveillance device associated with a phase is offline and the signal machine handles it by itself, if there are other video surveillance devices corresponding to other phases that have not been offline, the induction control mode can continue to run. If the video surveillance devices corresponding to other phases are all offline, the induction control mode is exited and the local fixed period control mode is entered.

[0099] For another example, when the target phase includes two phases in which the light state is green, the control operation of the traffic light by the signal machine can be determined according to the control signal values ​​of the two phases. Taking the target phase as east-west straight driving as an example, the east-west straight driving phase includes the east straight driving phase and the west straight driving phase. When the control signal values ​​corresponding to the east straight driving phase and the west straight driving phase are both the first preset values, the traffic light is controlled to stop the target phase, and when the control signal values ​​corresponding to the east straight driving phase and the west straight driving phase are both the second preset values, the traffic light is controlled to continue to maintain the target phase, etc.

[0100] In summary, in the induction control method provided by the present application, the operating data of the traffic lights at the target intersection at the current moment is sent to the edge computing device, so that the edge computing device can combine the video stream data collected by the existing video surveillance equipment at the target intersection to determine the control signal values ​​of each phase when the light state is green in the target stage, and then the traffic light can control the traffic flow at the target intersection according to the control signal values ​​of each phase, which can effectively improve the traffic efficiency of vehicles and the safety of pedestrians crossing the street, and has flexible control and low application cost.

[0101] Based on the same inventive concept as the aforementioned embodiment, the sensing control method provided in this embodiment is specifically described below through a specific example. In this example, the video surveillance device is used as the surveillance device and the edge computing device is used as the edge computing box.

[0102] The induction control method provided in this embodiment can be implemented by an induction control system, such as Figure 5 As shown, the system may include monitoring equipment, edge computing boxes and signal machines. The functions of each part of the system are described below:

[0103] Monitoring equipment: including but not limited to electric police, checkpoints, traffic cameras, and radar equipment, providing a stable video stream for the edge computing box.

[0104] The edge computing box includes six functional modules: video annotation, target detection, parameter configuration, light state reading, data analysis, and decision output. Among them, video annotation: first set any rectangle as the analysis range of the edge computing box in the video stream screen and select the motor vehicle or pedestrian recognition algorithm. Target detection: set any octagonal target detection area within the analysis range determined by the video annotation. If the motor vehicle algorithm is selected, a virtual import motor vehicle detection area and an export motor vehicle overflow detection area are provided to set and count the number of motor vehicles. Otherwise, a virtual pedestrian detection area is provided to set and count the number of pedestrians. Parameter configuration: Initialize the configuration of global parameters (including general parameters, phase overflow parameters, and stage termination delay output parameters, etc.) and local parameters (phase-time period parameters) based on the current operation timing plan, intersection channelization, and traffic organization. Light state reading: The edge computing box reads the message of the signal machine through the protocol to obtain the current operation plan number, stage number, and light state of each phase. Data analysis: Based on the current operation plan and target detection data, analyze whether the current stage should perform operations such as stage extension, stage termination, phase early segment, and stage control mode degradation and recovery. Decision output: Based on the results of data analysis, it is output to the signal machine in the form of phase number and phase signal value.

[0105] The signal machine includes functional modules such as scheme configuration, control mode setting, output light state, and acceptance of execution instructions. Scheme configuration: the phase setting, stage setting, scheme setting, time period setting, and plan setting required by the signal machine in the local fixed-cycle control mode. Control mode setting: at least provide local fixed-cycle control mode and local sensing control mode. Output light state: provide the edge computing box with the light state of the current running scheme number, stage number, and phase. Acceptance of execution instructions: the signal machine receives the phase number and phase signal value sent by the edge computing box, and executes operations such as phase extension, phase termination, phase early segment, and phase control mode downgrade and recovery.

[0106] Based on the above-mentioned induction control system, the induction control method provided in this embodiment includes the following steps:

[0107] S1. Configure the signal machine and run the local fixed-cycle control mode.

[0108] Specifically, based on the timing plan of the target intersection, the signal machine performs phase setting, stage setting, plan setting, time period setting and plan setting respectively. Among them, phase setting: configure the corresponding detector (i.e. monitoring equipment) number, minimum green, maximum green and unit green light parameters for each phase in the phase setting interface. Stage setting: configure the corresponding phase, yellow flash, full red, delay, early break and other parameters for each stage in the stage setting interface. Plan setting: add the target stages one by one in the plan setting interface and configure the corresponding green light duration parameters. Time period setting: configure the plan for each time period in the time period setting interface, and select the local fixed cycle control mode in the corresponding operation mode. Plan setting: set the scheduling plan in the plan setting interface. If it is the same every day, select Sunday to Saturday for the scheduling week.

[0109] S2. The edge computing box connects to the signal machine and video stream and initializes parameter configuration.

[0110] Among them, the edge computing box is connected to the signal machine, and the edge computing box and the signal machine can use a protocol connection to read the operation plan number, stage number and phase light state of the signal machine in real time. Take the north-south straight phase of the morning peak plan at time t as an example, as shown in Table 1.

[0111] Table 1

[0112]

[0113] Among them, the edge computing box also needs to access the video stream, set up a virtual detection area and generate detection data. The monitoring equipment connected to the edge computing box is not limited to electric police, checkpoints, video cameras and radar vision equipment. In this embodiment, the video stream of the monitoring equipment connected from the video terminal box is taken as an example. Specifically, the edge computing box first obtains the video stream from the video terminal box through the switch, and then selects the video stream to configure the target analysis area, and selects the motor vehicle recognition algorithm or the pedestrian algorithm, and then configures the target detection area in the target analysis area, and binds the associated phase, as shown in Table 2 below. The specific phase can be referred to Figure 1 shown.

[0114] Table 2

[0115]

[0116]

[0117] Among them, the parameter initialization configuration of the edge computing box includes global parameter initialization configuration and local phase-time period parameter initialization configuration. Among them, the global parameter initialization configuration is divided into global general parameter configuration (as shown in Table 3), phase overflow parameter configuration (as shown in Table 4) and stage delay output parameter configuration (as shown in Table 5).

[0118] Table 3

[0119] Serial number name meaning parameter unit 1 INT Video stream capture interval 1 Second 2 ANO The maximum number of consecutive times the video stream is offline 5 Second-rate 3 TJNO Overflow consecutive maximum number of times threshold 3 Second-rate 4 TJN Maximum continuous frequency threshold of the empty state in the stage 3 Second-rate

[0120] Table 4

[0121] serial number Phase number Name of exit inspection area Number of vehicles (unit: vehicle) 1 Phase 1 Turn left from the east exit 8 2 Phase 2 Go east straight 8 3 Phase 5 Exit West Left 8 4 Phase 6 West straight exit 8 5 Phase 9 Exit south left 8 6 Phase 10 Exit South 8 7 Phase 13 Exit North Left 8 8 Phase 14 North straight exit 8

[0122] Table 5

[0123] serial number name Duration (unit: seconds) 1 Go straight east-west 0 2 Turn left 0 3 Go straight north and south 0 4 Turn left from north to south 0 5 Dongxiang released 5 6 Western release 0 7 South released 0 8 North phase release 0

[0124] Among them, the phase-time parameter initialization configuration includes the configuration of the time period, traffic plan number, minimum green light duration, maximum green light duration and vehicle number threshold, as shown in Table 6.

[0125] Table 6

[0126]

[0127] Regarding the time period, the time period range configured by the edge computing box should cover the time range of the signal setting period. For example, the time period corresponding to the current morning peak timing plan is 7:30-8:30, and the corresponding edge computing time period range can be set to 7:00-9:00. In addition, crossover between time periods is allowed.

[0128] Regarding the operation logic, in order to adapt to the logical judgment of the number of pedestrians and motor vehicles and thresholds required for the termination of different periods, two logical operations, "sum" and "addition", are provided. Among them, "sum" is to judge the number of vehicles or pedestrians in the detection area of ​​each phase separately with the corresponding threshold; "addition" is to sum the number of vehicles in different detection areas of all phases in the same stage and judge with the sum of the corresponding thresholds, as shown in Table 7.

[0129] Table 7

[0130]

[0131] About the minimum green light duration of the phase j t min , which is the minimum green light duration of the phase under the signal control scheme in the current period.

[0132] About the maximum green light duration of the phase j t max , which is the maximum green light duration of the phase under the signal control scheme in the current period.

[0133] Regarding the threshold of the number of vehicles at the phase exit j NO, the vehicle number threshold set for the phase-bound exit detection zone.

[0134] Regarding the phase import vehicle number threshold jNJC, it is the vehicle number threshold set for the phase-bound import detection area.

[0135] About Phase Pedestrian Thresholdj NJP is the pedestrian number threshold set for the phase-bound pedestrian detection zone.

[0136] S3, edge computing box performs data processing and decision output.

[0137] Specifically, the edge computing box decides whether to end the current stage or phase based on the online status of the monitoring equipment, the overflow status of the exit, the number of vehicles in the import detection area, the number of pedestrians in the sidewalk detection area and the threshold value through data analysis and control algorithm, and then sends it to the signal machine for execution in the form of a phase signal value, including the following processes:

[0138] S31) captures the video stream every second to obtain the target detection area value, reads the signal message information and obtains the current time period operation logic value from the basic parameters.

[0139] Specifically, the target detection zone value is obtained, including the number of vehicles in the exit detection zone at time t and phase j. j no t , Number of vehicles in the import inspection area j njc t and the number of pedestrians in the sidewalk detection area j njp t ; Read the signal message information, including the signal operation plan number, stage number and phase light state of the signal at time t; obtain the current time period operation logic value, including taking the current time period operation logic value as "sum" or "addition".

[0140] S32) Enter stage p, record the current time as t=1, and calculate the minimum decision time of each stage p tp min and the maximum decision moment p tp max :

[0141] Minimum decision time p tp min : The minimum value of the minimum green light duration corresponding to all motor vehicle phases in phase p minus the delay duration corresponding to all motor vehicle phases j The maximum value in TD minus 4 seconds, that is p tp min =min{ j t min}-max{ j TD}-4.

[0142] Maximum decision moment in stage p tp max : The minimum value of the maximum green light duration corresponding to all motor vehicle phases in phase p minus the delay duration corresponding to all motor vehicle phases j The maximum value in TD minus 4 seconds, that is p tpmax =min{ j t max}-max{ j TD}-4.

[0143] S33) Calculate the temporary state value, that is, calculate the offline state value based on the real-time data and the relevant threshold value Overflow signal value And empty state value.

[0144] Offline status value If the video stream of the detection area corresponding to phase j at time t is offline, then the offline state value is recorded on the contrary,

[0145] Overflow status value If the number of vehicles in the overflow detection area (i.e., exit detection area) corresponding to phase j at time t is Greater than threshold j NO, then record the overflow status value on the contrary,

[0146] Empty status value, including import status value p tpc t and sidewalk status value p tpp t Two types.

[0147] Import status value p tpc t : When the operation logic is "and" at time t, if the number of vehicles in the import inspection area corresponding to each j phase of stage p at time t is Both are greater than the corresponding threshold of the number of imported vehicles Then record the import state value of stage p at time t p tpc t is 1, otherwise it is 0; when the operation logic is "add" at time t, if the number of vehicles in the import inspection area corresponding to each j phase of stage p at time t is The sum is greater than the corresponding threshold of the number of imported vehicles The sum of the import state value of stage p at time t is p tpc t is 1, otherwise it is 0.

[0148] Sidewalk status value p tpp t : When the operation logic is "and" at time t, if the number of pedestrians in the pedestrian detection area corresponding to each j phase of the p phase at time t is are greater than the corresponding pedestrian detection zone threshold Then record the sidewalk state value at time t and stage p as p tppt is 1, otherwise it is 0. When the operation logic is "add" at time t, if the number of pedestrians in the pedestrian detection area corresponding to each j phase of the p phase at time t The sum is greater than the corresponding pedestrian detection zone threshold The sum of the sidewalk state value at time t and stage p is p tpp t is 1, otherwise it is 0.

[0149] It should be noted that the j phase mentioned in this step is the phase when the light state is green.

[0150] S34) Calculate the decision signal value: respectively according to the minimum decision time not reached at time t p tp min , minimum decision time p tp min But it has not reached the moment of maximum decision p tp max and the moment of maximum decision p tp max Calculate offline signal values ​​in three situations Overflow signal value Empty signal value p op t and delayed signal value p od t .

[0151] Case 1: Time t does not reach the minimum decision time p tp min , that is, t< p tp min :

[0152] Offline signal value Record the offline signal value of phase j in phase p at time t

[0153] Overflow signal value Record the overflow signal value of phase j in phase p at time t

[0154] Empty signal value p op t :Record the empty discharge signal value of stage p at time t p op t =1;

[0155] Delayed signal value p od t :Record the delayed signal value of stage p at time t p od t =0.

[0156] Case 2: Exceeding the minimum decision timep tp min The biggest decision moment has not yet arrived p tp max ,Right now p tp min ≤t< p tp max .

[0157] Offline signal value If the video stream offline state value of the corresponding detection area at time t, stage p, and phase j The number of consecutive times reaches the maximum number of consecutive times of the video stream offline state ANO, that is Then record the offline signal value of phase j in stage p at time t as on the contrary,

[0158] Overflow signal value If the overflow state value of the exit detection zone bound to phase j in phase p at time t The number of consecutive times reaches the overflow maximum consecutive times threshold TJNO, that is, Record the overflow signal value of phase j in phase p at time t on the contrary,

[0159] Empty signal value p op t :If the vehicle empty state value of the import inspection area corresponding to the p stage at time t p tpc t =0 consecutive times reaches the maximum continuous frequency threshold of the stage empty state in stage p p TJN and pedestrian vacant status value p tpp t =0, that is and p tpp t =0, record the empty release signal value of stage p at time t p op t =0, otherwise, p op t =1.

[0160] Delayed signal value p od t : Delay is required due to factors such as the distance between the detection area and the parking line and the interference of pedestrians and non-motor vehicles at different times. p The TD termination phase ensures that the detected vehicle passes without stopping. p TD is the delay time corresponding to stage p. If the empty release signal value of stage p at time t2 is t2 is the p stage At the corresponding moment, time t2 may be before time t, or it may be after time t, that is, time t2 has not yet occurred, so the delayed signal value at time t is p od t satisfy:

[0161]

[0162] Scenario 3: The maximum decision moment has been reached p tp max , that is, t≥ p tp max .

[0163] Offline signal value The phase offline signal value of phase j in phase p at time t is

[0164] Overflow signal value Record the overflow signal value of phase j in phase p at time t

[0165] Empty signal value p op t :Record the empty discharge signal value of stage p at time t p op t =0;

[0166] Delayed signal value p od t :Record the delayed signal value of stage p at time t p od t =0.

[0167] S35) Output phase signal value And update the time t.

[0168] For the j phase in the p phase, determine the phase signal value of the j phase at time t-1 Is it equal to 0 or 3? If so, determine the phase signal value of the j phase at time t Take the phase signal value at time t-1 Otherwise, take the offline signal value Overflow signal value Empty signal value p op t and delayed signal value p od t Then, the process updates t=t+1 and returns to step S31.

[0169] S4) The operation mode of the signal machine is set to the local sensing control mode to receive and execute the message output by the edge computing box. If the current stage is a single-phase stage, proceed to step S41); otherwise, proceed to step S42).

[0170] S41) Single-phase stage: In the current stage, there is only a single motor vehicle phase, and the operations shown in Table 8 are performed according to the phase signal value (ie, code) in the message.

[0171] Table 8

[0172]

[0173] S42) Two-phase stage: There are two motor vehicle phases in the current stage, with a total of 16 combinations. Taking the stage including the south straight and north execution phases as an example, the corresponding phase signal value combination and instruction meaning are shown in Table 9.

[0174] Table 9

[0175]

[0176] In addition, global common parameters and local phase-period parameters can be tuned to optimize the control of edge computing.

[0177] In summary, in the induction control method provided in the above embodiment, based on the existing monitoring equipment such as electric alarms or checkpoints at the traffic light-controlled intersection, combined with the signal machine and equipped with edge computing equipment, the video stream is used as input, and the video analysis technology is used to count the number of incoming vehicles, the number of people on the sidewalk and the number of outgoing vehicles. Combined with the human-vehicle induction control method and the output of the phase signal value to the signal machine running in the local induction control mode, the phase extension, phase termination, phase early stage and phase control mode degradation and recovery of the timing plan are realized, which can greatly reduce the application cost of induction control and improve the safety of pedestrians crossing the street and the efficiency of vehicle traffic.

[0178] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the memory stores computer program instructions for execution on the processor, and when the processor executes the computer program instructions, the induction control method described above is implemented.

[0179] The embodiment of the present application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the above-mentioned induction control method.

Claims

1. A sensing control method, applied to edge computing equipment, characterized in that: The method comprises: Obtaining the operation data of the traffic lights at the target intersection sent by the signal machine at the current moment; the operation data includes the operation plan, the target stage of execution and the light state of each phase in the target stage; Acquire video stream data at a current moment collected by at least one video surveillance device disposed at a target intersection; According to the operation plan and the video stream data, determine the control signal value of each phase of the light state being green in the target phase; A control instruction including a control signal value for each phase is sent to the signal.

2. The method according to claim 1, characterized in that The operation scheme includes operation control logic and target object quantity thresholds corresponding to each phase in the target stage; According to the operation plan and the video stream data, the control signal value of each phase of the light state of the target stage is determined, including: Performing phase correlation detection on the video stream data to obtain detection results of the video stream sub-data associated with each phase in which the light state is green in the target phase; According to the operation control logic, the target object quantity threshold corresponding to each phase and the detection result of the video stream sub-data associated with each phase, the control signal value of each phase whose light state is green in the target stage is determined.

3. The method according to claim 2, characterized in that Perform phase correlation detection on the video stream data to obtain detection results of the video stream sub-data associated with each phase in which the light state is green in the target phase, including: Determine the target video stream sub-data associated with the target phase from the video stream data; the target phase is any phase in the target phase where the light state is green; The target objects in the target video stream sub-data that are in the target detection area corresponding to the target phase are identified and detected to obtain the number of target objects in the target detection area corresponding to the target phase.

4. The method according to claim 3, characterized in that According to the operation control logic, the target object quantity threshold corresponding to each phase, and the detection result of the video stream sub-data associated with each phase, the control signal value of each phase whose light state is green in the target stage is determined, including: According to the operation control logic and the target object quantity threshold corresponding to each phase in the target phase and the detection results of the video stream sub-data associated with each phase, the offline state value and the overflow state value of each phase with a green light in the target phase and the empty state value of the target phase are determined; According to the minimum decision time and maximum decision time of the target stage, the offline state value and overflow state value of each phase with the light state of green light in the target stage, and the idle state value of the target stage, determine the offline signal value and overflow signal value of each phase with the light state of green light in the target stage, as well as the idle signal value and delayed signal value of the target stage at the current moment; the minimum decision time and the maximum decision time are determined based on the minimum green light duration, the maximum green light duration and the output delay duration corresponding to all phases in the target stage respectively; According to the offline signal value and overflow signal value of each phase with the light state of green light in the target stage and the idle signal value and delayed signal value of the target stage at the current moment, the control signal value of each phase with the light state of green light in the target stage is determined.

5. The method according to claim 4, characterized in that The target objects include vehicles and pedestrians; the detection areas include the import vehicle detection area, the export vehicle detection area and the pedestrian detection area; the target object quantity threshold includes the import vehicle quantity threshold, the export vehicle quantity threshold and the pedestrian quantity threshold; the empty state value includes the import state value and the sidewalk state value; According to the operation control logic and the target object quantity threshold corresponding to each phase in the target stage and the detection result of the video stream sub-data associated with each phase, the offline state value and the overflow state value of each phase with the light state of green light in the target stage and the empty state value of the target stage are determined, including: If the detection result indicates that the video stream sub-data associated with the target phase is not obtained, then the offline state value of the target phase is marked as a first preset value, otherwise the offline state value of the target phase is marked as a second preset value; If the detection result indicates that the number of vehicles in the exit vehicle detection zone corresponding to the target phase is greater than the exit vehicle number threshold, the overflow state value of the target phase is marked as a first preset value, otherwise the overflow state value of the target phase is marked as a second preset value; When the operation control logic is the first operation logic, if the detection result indicates that the number of vehicles in the import vehicle detection area corresponding to each phase in which the light state is green in the target stage is respectively greater than the corresponding import vehicle number threshold, then the import state value of the target stage is marked as the first preset value, otherwise the import state value of the target stage is marked as the second preset value; if the detection result indicates that the number of pedestrians in the pedestrian detection area corresponding to each phase in which the light state is green in the target stage is respectively greater than the corresponding pedestrian number threshold, then the sidewalk state value of the target stage is marked as the first preset value, otherwise the sidewalk state value of the target stage is marked as the second preset value; When the operation control logic is the second operation logic, if the detection result indicates that the sum of the number of vehicles in the import vehicle detection area corresponding to each phase in which the light state is green in the target stage is greater than the sum of the corresponding import vehicle number thresholds, then the import state value of the target stage is marked as the first preset value, otherwise the import state value of the target stage is marked as the second preset value; if the detection result indicates that the sum of the number of pedestrians in the pedestrian detection area corresponding to each phase in which the light state is green in the target stage is greater than the sum of the corresponding pedestrian number thresholds, then the sidewalk state value of the target stage is marked as the first preset value, otherwise the sidewalk state value of the target stage is marked as the second preset value.

6. The method according to claim 4 or 5, characterized in that According to the minimum decision time and the maximum decision time of the target phase, the offline state value and the overflow state value of each phase with the light state of green light in the target phase, and the empty release state value of the target phase, the offline signal value and the overflow signal value of each phase with the light state of green light in the target phase, and the empty release signal value and the delayed signal value of the target phase at the current moment are determined, including: If the minimum decision time of the target stage is not reached at the current moment, the offline signal value and the overflow signal value of each phase whose light state is green in the target stage are determined to be the second preset value and the empty release signal value and the delay signal value of the target stage at the current moment are determined to be the first preset value and the second preset value; If the current moment has reached the minimum decision moment of the target stage but has not reached the maximum decision moment of the target stage, then when the number of consecutive times that the offline state value of the target phase is the first preset value reaches the preset maximum offline number, the offline signal value of the target phase is marked as the third preset value, otherwise the offline signal value of the target phase is marked as the second preset value; when the number of consecutive times that the overflow state value of the target phase is the first preset value reaches the preset maximum overflow number, the overflow signal value of the target phase is marked as the fourth preset value, otherwise the overflow signal value of the target phase is marked as the second preset value; when the number of consecutive times that the import state value of the target stage is the second preset value reaches the preset maximum empty release number and the sidewalk state value is the second preset value, the empty release signal value of the target stage at the current moment is marked as the second preset value, otherwise the empty release signal value of the target stage at the current moment is marked as the first preset value; when the difference between the current moment and the start moment when the empty release signal value of the target stage is the second preset value is less than or equal to the preset delay time, the delay signal value of the target stage at the current moment is marked as the first preset value, otherwise the delay signal value of the target stage at the current moment is marked as the second preset value; If the maximum decision moment of the target stage has been reached at the current moment, the offline signal value and overflow signal value of each phase with the light state of green light in the target stage are determined to be the second preset values ​​respectively, and the idle signal value and delayed signal value of the target stage at the current moment are determined to be the second preset values ​​respectively.

7. The method according to claim 6, characterized in that According to the offline signal value and overflow signal value of each phase whose light state is green in the target phase and the empty release signal value and delayed signal value of the target phase at the current moment, the control signal value of each phase whose light state is green in the target phase is determined, including: Detecting whether the control signal value of the target phase at the previous moment is a preset signal value; If yes, the preset signal value is used as the control signal value of the target phase at the current moment; If not, the maximum value among the offline signal value, overflow signal value of the target phase at the current moment, the empty signal value and the delayed signal value of the target phase at the current moment is used as the control signal value of the target phase at the current moment.

8. An induction control method, applied to a signal machine, characterized in that: include: Send the current operation data of the traffic lights at the target intersection to the edge computing device; the operation data includes the operation plan, the target stage of execution, and the light state of each phase in the target stage; Obtaining a control instruction sent by the edge computing device including a control signal value of each phase in which the light state is green in the target phase; Control the traffic lights at the target intersection to execute control instructions.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores computer program instructions for being executed on the processor, and when the processor executes the computer program instructions, the induction control method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium, characterized in that: It stores computer instructions, and when the computer instructions are executed by a processor, the induction control method according to any one of claims 1 to 8 is implemented.

Images