Millimeter-wave radar target detection method for low false alarm processing in complex environments

Abstract

The present invention relates to a millimeter wave radar target detection method for low false alarm processing in a complex environment, comprising: after the millimeter wave radar receives the elevator car working mode instruction, it determines whether background initialization processing is required; enters the background learning mode to ensure that the background of the detection area inside and outside the elevator door is clean and there are no moving targets, and collects the point cloud information of each target under the current background; or directly enters the normal working mode, and starts to collect the target point cloud information after receiving the signal that the elevator door is opened; the collected point cloud information is preliminarily screened according to the set ROI threshold, and the target point cloud number set that meets the threshold is screened out for each frame; the obtained target point cloud number set is averaged by sliding, and the corresponding mode selection processing is performed. The present invention also relates to a corresponding device, a processor and a storage medium thereof, which can quickly and accurately perceive the target and its moving trend in a complex environment, with a low false alarm rate, thereby improving the safety of elevator operation.

Description

Technical Field

[0001] The present invention relates to the field of millimeter wave radar target detection, in particular to the field of elevator door sensing, and specifically refers to a millimeter wave radar target detection method, device, processor and computer-readable storage medium thereof for low false alarm processing in a complex environment. Background Art

[0002] In modern buildings, automatic door-controlled elevators have become an indispensable means of vertical transportation. The elevator door sensor is an important part of the elevator system. It can realize the automatic opening and closing of the elevator door, providing passengers with a convenient entry and exit experience. Among them, how to quickly sense the status of passengers inside and outside the elevator door, prevent passengers from being pinched, and ensure passenger safety is an important role that the elevator door sensor should play.

[0003] Currently, most elevator door sensors use infrared light curtains, such as patent documents CN201406257Y and CN219906645U. However, the sensing distance and coverage of infrared light curtain sensors are limited. They can only cover and sense the narrow area between the two doors and cannot sense the movement trend of passengers or objects inside and outside the door. And they may be affected by ambient light, so it is impossible to avoid passengers or objects being pinched, posing a safety hazard. There are also TOF laser radars used as sensors, such as patent document CN117401536A, but TOF laser radars can only measure the distance to the target and cannot determine the actual movement state of the target. When a moving target crosses in front of the elevator door, false alarms are likely to occur, causing the elevator door to be unable to close.

[0004] In recent years, due to the continuous development of millimeter-wave radar technology, millimeter-wave radar has also been used in elevator door sensing. However, since most of the materials of elevator cars and elevator doors are metal materials, there is a strong reflection of electromagnetic waves, which makes the use environment of millimeter-wave radar complex and prone to false alarms, resulting in abnormal opening and closing of elevator doors. Patent document CN117228472A uses a combination of camera and millimeter-wave radar to determine whether there are passengers waiting to board. However, it can only detect whether there are passengers waiting in front of the door, and cannot determine the target movement trajectory, such as pedestrians crossing, resulting in false alarms. Moreover, the system is complex and contains two types of sensors, and the application cost is high. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a millimeter wave radar target detection method, device, processor and computer-readable storage medium thereof for low false alarm processing in a complex environment.

[0006] In order to achieve the above-mentioned object, the millimeter wave radar target detection method, device, processor and computer-readable storage medium thereof for low false alarm processing in a complex environment of the present invention are as follows:

[0007] The main feature of the millimeter wave radar target detection method for low false alarm processing in a complex environment is that the method comprises the following steps:

[0008] (1) After the millimeter-wave radar receives the elevator car working mode instruction, it determines whether background initialization processing is required;

[0009] (2) Enter the background learning mode, ensure that the background of the detection area inside and outside the elevator door is clean and there are no moving targets, and collect the point cloud information of each target under the current background; or directly enter the normal working mode, and start collecting the target point cloud information when receiving the signal that the elevator door is open;

[0010] (3) Preliminary screening of the collected point cloud information is performed according to the set ROI threshold, and the target point cloud sets that meet the threshold are selected for each frame;

[0011] (4) Take sliding average of the acquired target point cloud set and select the following mode;

[0012] (5) In the background learning mode, the elevator door opening threshold is directly obtained, and the normal working mode is entered after the background initialization processing is completed; or

[0013] In the normal working mode, it is necessary to determine whether the point cloud number of the current frame is greater than the elevator door opening threshold, and whether the target movement flag of the current frame is greater than the target movement threshold, and execute the corresponding working signal of the millimeter wave radar according to the judgment result to control the opening and closing of the elevator door in real time.

[0014] Preferably, the millimeter wave radar performs the following processing in the background learning mode:

[0015] After receiving the elevator car working mode instruction, the millimeter wave radar transmits a signal and receives an echo signal, obtains an intermediate frequency signal after mixing, and parses the point cloud information of each target in the current background from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , the information of each target point is recorded as target ( R , V , azi , elv , P );

[0016] Then in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, ,total frame; Indicates j The first frame in the original space target point cloud set i The information of the target point cloud, ,total Target point cloud.

[0017] Preferably, the step (3) is specifically as follows:

[0018] Set the elevator door sensor sensing range to ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio;

[0019] According to the setting , filter out the point cloud in each frame that meets the threshold range, called the spatial background target point cloud set, recorded as ,in, is the i-th target point cloud information of the j-th frame that meets the threshold range, ,total target point clouds; thus, the background target point cloud set that meets the threshold in each frame during the opening and closing process of the elevator door is obtained, which is recorded as ,in, , Indicates the number of background target point clouds that meet the threshold range in the jth frame;

[0020] ;

[0021] in, , .

[0022] Preferably, the step (4) is specifically as follows:

[0023] The obtained background target point cloud set is averaged by sliding in the following way, where the sliding window size is ,

[0024] ;

[0025] in, Represents the number of background target point clouds in each frame after sliding average, , Indicates j The number of background target point clouds in each frame after frame sliding averaging.

[0026] Preferably, in the background learning mode, step (5) is specifically as follows:

[0027] set up is the elevator door opening threshold, then

[0028] ;

[0029] in, represents the threshold coefficient;

[0030] At this point, the elevator door opening threshold value is obtained , completing the background initialization process.

[0031] Preferably, the millimeter wave radar performs the following processing in a normal working mode:

[0032] After receiving the elevator car working mode instruction, the millimeter wave radar transmits a signal and receives an echo signal, obtains an intermediate frequency signal after mixing, and parses the point cloud information of each target in the current background from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , the information of each target point is recorded as target ( R , V , azi , elv , P );

[0033] Then in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, Indicates j The first frame in the original space target point cloud set i The information of the target point cloud, ,total Target point cloud.

[0034] Preferably, the step (3) is specifically as follows:

[0035] Set the elevator door sensor sensing range to ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio;

[0036] Filter out the point cloud in each frame that meets the threshold range, which is called the target point cloud set and is recorded as ,in, For the j The frame that meets the threshold range i Target point cloud information, ,total target point clouds; thus, the set of scene target point clouds that meet the threshold in each frame during the opening and closing process of the elevator door is obtained, which is recorded as , Indicates the number of scene target point clouds that meet the threshold range in the jth frame;

[0037] ;

[0038] in, , .

[0039] Preferably, the step (4) is specifically as follows:

[0040] The obtained current frame scene target point cloud number and the stored historical frame scene target point cloud number set are averaged by sliding in the following way, where the sliding window size is ,

[0041] ;

[0042] in, Indicates the number of target point clouds in the current frame scene after sliding average.

[0043] Preferably, in the normal working mode, the step (5) comprises:

[0044] According to the target point cloud Information, calculate the center of gravity position of the target point cloud in the current frame ;in, , ; This forms the moving target center of gravity position set ,in j Indicates j frame;

[0045] The moving target gravity center moving mark set is ,and

[0046] ;

[0047] Get the center of gravity movement mark of the current frame scene target And the target center of gravity movement mark of the stored historical frame scene The sliding average is performed in the following manner, where the sliding window size is ;

[0048] .

[0049] Preferably, the step (5) further comprises:

[0050] like > and > , it is judged that someone is approaching the elevator door, and the radar outputs a door opening signal; if ≤ or ≤ , it is judged that no one is approaching the elevator door, and the radar delays a certain time to output the door closing signal, where The elevator door opening threshold. For moving targets close to the threshold;

[0051] When the millimeter-wave radar receives the signal that the elevator door is completely closed, it stops working.

[0052] The main features of the millimeter wave radar target detection device for low false alarm processing in complex environments are that the device comprises:

[0053] a processor configured to execute computer executable instructions;

[0054] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the millimeter wave radar target detection method for low false alarm processing in a complex environment are implemented.

[0055] The main feature of the radar-visual fusion target perception processor based on the hybrid strategy is that the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the steps of the millimeter-wave radar target detection method for low false alarm processing in a complex environment are implemented;

[0056] The computer-readable storage medium has the main feature of storing a computer program thereon, and the computer program can be executed by a processor to implement the steps of the millimeter-wave radar target detection method for low false alarm processing in a complex environment described above.

[0057] The millimeter-wave radar target detection method, device, processor and computer-readable storage medium for low false alarm processing in the complex environment of the present invention can well solve the problems of short sensing distance, limited sensing coverage range, high false alarm rate of TOF laser radar and millimeter-wave radar, etc. of current elevator door infrared light curtain and other sensors. The algorithm is simple and efficient, and can quickly and accurately sense the target and its movement trend in a complex environment with a low false alarm rate; it makes the elevator door sensing distance longer, the sensing range larger, and the sensing accuracy higher. It effectively avoids the elevator door from pinching people or objects, and improves the safety of elevator operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the structure of a specific embodiment of the present invention.

[0059] Figure 2 The present invention is a flow chart of the millimeter wave radar target detection method for low false alarm processing in the complex environment.

[0060] Figure 3 A schematic diagram of a background target point cloud data set that meets a threshold in each frame during the opening and closing process of an elevator door when background initialization is performed in an embodiment of the present invention.

[0061] Figure 4 When the background is initialized in an embodiment of the present invention, a schematic diagram of a background target point cloud set of each frame after sliding averaging is shown during the opening and closing process of the elevator door.

[0062] Figure 5 When the embodiment of the present invention is working normally, a moving target appears during the opening and closing process of the elevator door, and a schematic diagram of a scene target point cloud set that meets the threshold in each frame.

[0063] Figure 6 When the embodiment of the present invention is working normally, a moving target appears during the opening and closing process of the elevator door, and a schematic diagram of a target point cloud set of each frame scene after sliding averaging is shown.

[0064] Figure 7When the embodiment of the present invention works normally, a moving target appears during the opening and closing process of the elevator door, and a schematic diagram of the door opening and closing signal output by the radar.

[0065] Reference numerals

[0066] 1 4D millimeter wave radar (sensor)

[0067] 2 Elevator Doors

[0068] 3 Elevator Car DETAILED DESCRIPTION

[0069] In order to more clearly describe the technical content of the present invention, further description is given below in conjunction with specific embodiments.

[0070] Before describing in detail embodiments according to the present invention, it should be noted that, hereinafter, the terms "comprises", "includes" or any other variations are intended to cover non-exclusive inclusion, whereby a process, method, article or apparatus comprising a series of elements includes not only these elements, but also other elements not explicitly listed or inherent to such process, method, article or apparatus.

[0071] The millimeter wave radar target detection method for low false alarm processing in the complex environment, wherein the method comprises the following steps:

[0072] (1) After the millimeter-wave radar receives the elevator car working mode instruction, it determines whether background initialization processing is required;

[0073] (2) Enter the background learning mode, ensure that the background of the detection area inside and outside the elevator door is clean and there are no moving targets, and collect the point cloud information of each target under the current background; or directly enter the normal working mode, and start collecting the target point cloud information when receiving the signal that the elevator door is opened;

[0074] (3) Preliminary screening of the collected point cloud information is performed according to the set ROI threshold, and the target point cloud sets that meet the threshold are selected for each frame;

[0075] (4) Take sliding average of the acquired target point cloud set and select the following mode;

[0076] (5) In the background learning mode, the elevator door opening threshold is directly obtained, and the normal working mode is entered after the background initialization processing is completed; or

[0077] In the normal working mode, it is necessary to determine whether the point cloud number of the current frame is greater than the elevator door opening threshold, and whether the target movement flag of the current frame is greater than the target movement threshold, and execute the corresponding working signal of the millimeter wave radar according to the judgment result to control the opening and closing of the elevator door in real time.

[0078] As a preferred embodiment of the present invention, the millimeter wave radar performs the following processing in the background learning mode:

[0079] After receiving the elevator car working mode instruction, the millimeter wave radar transmits a signal and receives an echo signal, obtains an intermediate frequency signal after mixing, and parses the point cloud information of each target in the current background from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , the information of each target point is recorded as target ( R , V , azi , elv , P );

[0080] Then in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, ,total frame; Indicates j The first frame in the original space target point cloud set i The information of the target point cloud, ,total Target point cloud.

[0081] As a preferred embodiment of the present invention, the step (3) is specifically as follows:

[0082] Set the elevator door sensor sensing range to ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio;

[0083] According to the setting , filter out the point cloud in each frame that meets the threshold range, called the spatial background target point cloud set, recorded as ,in, is the i-th target point cloud information of the j-th frame that meets the threshold range, ,total target point clouds; thus, the background target point cloud set that meets the threshold in each frame during the opening and closing process of the elevator door is obtained, which is recorded as ,in, , Indicates the number of background target point clouds that meet the threshold range in the jth frame;

[0084] ;

[0085] in, , .

[0086] As a preferred embodiment of the present invention, the step (4) is specifically as follows:

[0087] The obtained background target point cloud set is averaged by sliding in the following way, where the sliding window size is ,

[0088] ;

[0089] in, Represents the number of background target point clouds in each frame after sliding average, , Indicates j The number of background target point clouds in each frame after frame sliding averaging.

[0090] As a preferred embodiment of the present invention, step (5) in the background learning mode is specifically as follows:

[0091] set up is the elevator door opening threshold, then

[0092] ;

[0093] in, represents the threshold coefficient;

[0094] At this point, the elevator door opening threshold value is obtained , completing the background initialization process.

[0095] As a preferred embodiment of the present invention, the millimeter wave radar performs the following processing in the normal working mode:

[0096] After receiving the elevator car working mode instruction, the millimeter wave radar transmits a signal and receives an echo signal, obtains an intermediate frequency signal after mixing, and parses the point cloud information of each target in the current background from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , the information of each target point is recorded as target ( R , V , azi , elv , P );

[0097] Then in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, Indicates j The first frame in the original space target point cloud set i The information of the target point cloud, ,total Target point cloud.

[0098] As a preferred embodiment of the present invention, the step (3) is specifically as follows:

[0099] Set the elevator door sensor sensing range to ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio;

[0100] Filter out the point cloud in each frame that meets the threshold range, which is called the target point cloud set and is recorded as ,in, For the j The frame that meets the threshold range iTarget point cloud information, ,total target point clouds; thus, the set of scene target point clouds that meet the threshold in each frame during the opening and closing process of the elevator door is obtained, which is recorded as , Indicates the number of scene target point clouds that meet the threshold range in the jth frame;

[0101] ;

[0102] in, , .

[0103] As a preferred embodiment of the present invention, the step (4) is specifically as follows:

[0104] The obtained current frame scene target point cloud number and the stored historical frame scene target point cloud number set are averaged by sliding in the following way, where the sliding window size is ,

[0105] ;

[0106] in, Indicates the number of target point clouds in the current frame scene after sliding average.

[0107] As a preferred embodiment of the present invention, in the normal working mode, step (5) includes:

[0108] According to the target point cloud Information, calculate the center of gravity position of the target point cloud in the current frame ;in, , ; This forms the moving target center of gravity position set ,in j Indicates j frame;

[0109] The moving target gravity center moving mark set is ,and

[0110] ;

[0111] Get the center of gravity movement mark of the current frame scene target And the target center of gravity movement mark of the stored historical frame scene The sliding average is performed in the following manner, where the sliding window size is ;

[0112] .

[0113] As a preferred embodiment of the present invention, the step (5) further comprises:

[0114] like > and > , it is judged that someone is approaching the elevator door, and the radar outputs a door opening signal; if ≤ or ≤ , it is judged that no one is approaching the elevator door, and the radar delays a certain time to output the door closing signal, where The elevator door opening threshold. For moving targets close to the threshold;

[0115] When the millimeter-wave radar receives the signal that the elevator door is completely closed, it stops working.

[0116] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0117] In a specific embodiment of the present invention, a 4D millimeter wave radar is installed in the elevator car near the upper part of the elevator door. The radar normal direction is parallel to the elevator door and perpendicular to the ground, the radar pitch direction is the elevator door opening and closing direction, and the radar azimuth direction is perpendicular to the elevator door.

[0118] The specific implementation steps are as follows:

[0119] 1. After the radar receives the background initialization signal, it enters the background initialization working mode. At this time, the background of the detection area inside and outside the elevator door needs to be clean, and there is no moving target except the normal opening and closing of the elevator door.

[0120] 2. The radar transmits a signal and receives an echo signal, and after mixing, an intermediate frequency signal is obtained. The point cloud information of each target in the current background is parsed from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , record the information of each target point as target ( R , V , azi , elv , P ). Therefore, in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, ,total frame. Indicates j The first frame in the original space target point cloud seti The information of the target point cloud, ,total The elevator door opens and closes five times freely when the radar background is initialized.

[0121] 3. Such as Figure 1 As shown in the figure, the shadow part is the projection of the elevator door sensor sensing area on the ground. The sensing range of the elevator door sensor is ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio. In this embodiment, ROI1 is set to (-3, 3, -2, 2, -55, 55, -30, 30, -2, 10).

[0122] 4. According to the set ROI, filter out the point cloud in each frame that meets the threshold range, which is called the spatial background target point cloud set and is recorded as .in, is the i-th target point cloud information of the j-th frame that meets the threshold range, ,total Thus, we get the background target point cloud set that meets the threshold in each frame during the opening and closing process of the elevator door, which is recorded as ,in, , Indicates the number of background target point clouds that meet the threshold range in the jth frame. Figure 3 shown.

[0123] ;

[0124] in, , .

[0125] 5. Perform sliding averaging on the background target point cloud set obtained, where the sliding window size is meanNum=20.

[0126] ;

[0127] in, Represents the number of background target point clouds in each frame after sliding average, , Represents the number of background target point clouds in each frame after the j-th frame sliding average. Figure 4 shown.

[0128] 6. Remember The elevator door opening threshold.

[0129] ;

[0130] in, represents the threshold coefficient, which is 1.2 in this embodiment.

[0131] At this point, the elevator door opening threshold value in this embodiment is obtained It is 6.18, and the background initialization is completed.

[0132] 7. When the radar receives the normal working mode signal, the radar enters the normal working mode.

[0133] 8. The radar is in normal working mode. When it receives the signal that the elevator door is open, the radar starts working.

[0134] 9. The radar transmits a signal and receives an echo signal, and after mixing, an intermediate frequency signal is obtained. The point cloud information of each target in the current scene is parsed from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , record the information of each target point as target ( R , V , azi , elv , P ). Therefore, in the current context, the original spatial target point cloud set is ,in, Represents the original space target point cloud set of the jth frame. Represents the information of the i-th target point cloud in the original space target point cloud set of the j-th frame, ,total Target point cloud.

[0135] 10. In this embodiment, ROI2 is set to (-3, 3, -2, 2, -55, 55, -30, 30, -0.5, 10). According to the set ROI2, the point cloud in each frame that meets the threshold range is screened out, which is called the target point cloud set and is recorded as .in, is the i-th target point cloud information of the j-th frame that meets the threshold range, ,total Thus, we get the set of scene target point clouds that meet the threshold in each frame during the opening and closing process of the elevator door, which is recorded as , Indicates the number of scene target point clouds that meet the threshold range in the jth frame. Figure 5 shown.

[0136] ;

[0137] in, , .

[0138] 11. Take a sliding average of the obtained current frame scene target point cloud number and the stored historical frame scene target point cloud number set. The sliding window size meanNum1 is 20.

[0139] ;

[0140] in, Indicates the number of scene target point clouds in the current frame (current moment) after sliding average. Figure 6 shown.

[0141] 12. According to the target point cloud Information, calculate the center of gravity position of the target point cloud in the current frame .

[0142] in: , .

[0143] This forms the moving target center of gravity position set , where j represents the jth frame.

[0144] 13. The target center of gravity movement mark set is ,

[0145] .

[0146] 14. Get the center of gravity of the scene target in the current frame and move it And the target center of gravity movement mark of the stored historical frame scene Perform sliding averaging, where the sliding window size meanNum2 is 10.

[0147] .

[0148] 15. If >6.18 and >1.5, it is judged that someone is approaching the elevator door, and the radar outputs a door opening signal. ≤6.18 or ≤1.5, it is judged that no one is approaching the elevator door, and the radar delays a certain period of time to output the door closing signal. Figure 6 and Figure 7 As shown, Figure 7 1 means that a moving target is detected and the radar sends a door-opening signal, and 0 means that no moving target is detected and the radar sends a door-closing signal.

[0149] 16. When the radar receives the signal that the elevator door is completely closed, the radar stops working.

[0150] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0151] It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device.

[0152] A person of ordinary skill in the art may understand that all or part of the steps of the method for implementing the above-mentioned embodiment may be completed by instructing the relevant hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.

[0153] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0154] In the description of this specification, the description with reference to the terms "an embodiment", "some embodiments", "example", "specific example", or "embodiment" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0155] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

[0156] The millimeter-wave radar target detection method, device, processor and computer-readable storage medium for low false alarm processing in the complex environment of the present invention can well solve the problems of short sensing distance, limited sensing coverage range, high false alarm rate of TOF laser radar and millimeter-wave radar, etc. of current elevator door infrared light curtain and other sensors. The algorithm is simple and efficient, and can quickly and accurately sense the target and its movement trend in a complex environment with a low false alarm rate; it makes the elevator door sensing distance longer, the sensing range larger, and the sensing accuracy higher. It effectively avoids the elevator door from pinching people or objects, and improves the safety of elevator operation.

[0157] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it is apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Therefore, the specification and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A millimeter wave radar target detection method for low false alarm processing in a complex environment, characterized in that: The method comprises the following steps: (1) After the millimeter-wave radar receives the elevator car working mode instruction, it determines whether background initialization processing is required; (2) Enter the background learning mode, ensure that the background of the detection area inside and outside the elevator door is clean and there are no moving targets, and collect the point cloud information of each target under the current background; or directly enter the normal working mode, and start collecting the target point cloud information when receiving the signal that the elevator door is open; (3) Preliminary screening of the collected point cloud information is performed according to the set ROI threshold, and the target point cloud sets that meet the threshold are selected for each frame; (4) Take sliding average of the acquired target point cloud set and select the following mode; (5) In the background learning mode, the elevator door opening threshold is directly obtained, and the normal working mode is entered after the background initialization processing is completed; or In the normal working mode, it is necessary to determine whether the point cloud number of the current frame is greater than the elevator door opening threshold, and whether the target movement flag of the current frame is greater than the target movement threshold, and execute the corresponding working signal of the millimeter wave radar according to the judgment result to control the opening and closing of the elevator door in real time.

2. The millimeter wave radar target detection method for low false alarm processing in a complex environment according to claim 1, characterized in that: The millimeter wave radar performs the following processing in the background learning mode: After receiving the elevator car working mode instruction, the millimeter wave radar transmits a signal and receives an echo signal, obtains an intermediate frequency signal after mixing, and parses the point cloud information of each target in the current background from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , the information of each target point is recorded as target ( R , V , azi , elv , P ); Then in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, ,total frame; Indicates j The first frame in the original space target point cloud set i The information of the target point cloud, ,total Target point cloud.

3. The millimeter wave radar target detection method for low false alarm processing in a complex environment according to claim 2, characterized in that: The step (3) is specifically as follows: Set the elevator door sensor sensing range to ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio; According to the setting , filter out the point cloud in each frame that meets the threshold range, called the spatial background target point cloud set, recorded as ,in, is the i-th target point cloud information of the j-th frame that meets the threshold range, ,total target point clouds; thus, the background target point cloud set that meets the threshold in each frame during the opening and closing process of the elevator door is obtained, which is recorded as ,in, , Indicates the number of background target point clouds that meet the threshold range in the jth frame. The specific screening method is as follows: ； in, , .

4. The millimeter wave radar target detection method for low false alarm processing in a complex environment according to claim 3 is characterized in that: The step (4) is specifically as follows: The obtained background target point cloud set is averaged by sliding in the following way, where the sliding window size is , ； in, Represents the number of background target point clouds in each frame after sliding average, , Indicates j The number of background target point clouds in each frame after frame sliding averaging.

5. The millimeter wave radar target detection method for low false alarm processing in a complex environment according to claim 4, characterized in that: In the background learning mode, step (5) is specifically as follows: set up is the elevator door opening threshold, then ； in, represents the threshold coefficient; At this point, the elevator door opening threshold value is obtained , completing the background initialization process.

6. The millimeter wave radar target detection method for low false alarm processing in a complex environment according to claim 1, characterized in that: The millimeter wave radar performs the following processing in the normal working mode: After receiving the elevator car working mode instruction, the millimeter wave radar transmits a signal and receives an echo signal, obtains an intermediate frequency signal after mixing, and parses the point cloud information of each target in the current background from the intermediate frequency signal, including: radial distance R , radial velocity V , azimuth azi , Pitch angle elv , signal-to-noise ratio P , the information of each target point is recorded as target ( R , V , azi , elv , P ); Then in the current context, the original spatial target point cloud set is ,in, Indicates j Frame original space target point cloud set, Indicates j The first frame in the original space target point cloud set i The information of the target point cloud, ,total Target point cloud.

7. The millimeter wave radar target detection method for low false alarm processing in complex environment according to claim 6, characterized in that: The step (3) is specifically as follows: Set the elevator door sensor sensing range to ,in, Indicates the X-axis distance of the sensor range outside the elevator door projected on the ground. Indicates the X-axis distance of the sensor range inside the elevator door projected on the ground. Indicates the positive Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates the negative Y-axis distance of the elevator door's inner / outer sensing range projected on the ground. Indicates that the target is at the upper limit of the radar's azimuth angle. Indicates that the target is at the lower limit of the radar's azimuth angle. Indicates that the target is at the upper limit of the radar's elevation angle. Indicates that the target is at the lower limit of the radar's elevation angle. represents the upper limit of the target radial velocity, Indicates the upper limit of the target signal-to-noise ratio; Filter out the point cloud in each frame that meets the threshold range, which is called the target point cloud set and is recorded as ,in, For the j The frame that meets the threshold range i Target point cloud information, ,total target point clouds; thus, the set of scene target point clouds that meet the threshold in each frame during the opening and closing process of the elevator door is obtained, which is recorded as , Indicates the number of scene target point clouds that meet the threshold range in the jth frame. The specific screening method is as follows: ； in, , .

8. The millimeter wave radar target detection method for low false alarm processing in complex environment according to claim 7, characterized in that: The step (4) is specifically as follows: The obtained current frame scene target point cloud number and the stored historical frame scene target point cloud number set are averaged by sliding in the following way, where the sliding window size is , ； in, Indicates the number of target point clouds in the current frame scene after sliding average.

9. The millimeter wave radar target detection method for low false alarm processing in complex environment according to claim 8, characterized in that: In the normal working mode, step (5) includes: According to the target point cloud Information, calculate the center of gravity position of the target point cloud in the current frame ;in, , ; This forms the moving target center of gravity position set ,in j Indicates j frame; The moving target gravity center moving mark set is ,and ； Get the center of gravity movement mark of the current frame scene target And the target center of gravity movement mark of the stored historical frame scene The sliding average is performed in the following manner, where the sliding window size is ; 。 10. The millimeter wave radar target detection method for low false alarm processing in complex environment according to claim 9, characterized in that: The step (5) further includes: like > and > , it is judged that someone is approaching the elevator door, and the radar outputs a door opening signal; if ≤ or ≤ , it is judged that no one is approaching the elevator door, and the radar delays a certain time to output the door closing signal, where The elevator door opening threshold. For moving targets close to the threshold; When the millimeter-wave radar receives the signal that the elevator door is completely closed, it stops working.

11. A millimeter wave radar target detection device for low false alarm processing in a complex environment, characterized in that: The device comprises: a processor configured to execute computer executable instructions; A memory stores one or more computer executable instructions, and when the computer executable instructions are executed by the processor, the steps of the millimeter wave radar target detection method for low false alarm processing in a complex environment according to any one of claims 1 to 10 are implemented.

12. A radar-visual fusion target perception processor based on a hybrid strategy, characterized in that: The processor is configured to execute computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the millimeter wave radar target detection method for low false alarm processing in a complex environment described in any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program can be executed by a processor to implement the steps of the millimeter wave radar target detection method for low false alarm processing in a complex environment in any one of claims 1 to 10.

Images