CN114241015B - Method and device for counting number of people who pass in and out based on millimeter wave radar - Google Patents

Abstract

The invention discloses a method for counting the number of people passing in and out based on a millimeter wave radar, which comprises the steps of carrying out point cloud cluster tracking processing on a human target and outputting an output clustered target set; predicting and updating each target in the clustered target set by adopting an extended Kalman tracking algorithm, and acquiring the position information of the target in real time; converting the target position information into a Cartesian coordinate system from a polar coordinate to obtain a tracking target set; continuously tracking the same track ID, judging the in-out state of the tracked target according to the position and the speed of the tracked target relative to the first crossing line, the second crossing line and the third crossing line, adding 1 to the number of the in-out persons of the target in the tracked state, and subtracting 1 from the number of the in-out persons of the target in the tracked state. The method allocates a unique track ID to the pedestrian in the monitoring area, and judges the position and the speed direction on the adjacent frames; effectively avoiding repeated counting of people staying at the doorway.

Description

Method and device for counting number of people who pass in and out based on millimeter wave radar

Technical Field

The invention belongs to the technical field of monitoring, and particularly relates to a method and a device for counting the number of people passing in and out based on a millimeter wave radar.

Background

The existing statistical method for the number of people passing in and out comprises an image processing technology and an infrared correlation technology. The image processing is based on optical camera imaging, and algorithms such as target extraction, target recognition, track analysis and the like are mainly adopted; it is greatly influenced by ambient light, and has complex treatment process and high cost. The infrared correlation adopts a transmitting head and a receiving head, if an object passes through the middle, light rays are shielded, a pulse signal is output to trigger a counting circuit, the transmitting head and the receiving head are combined into an infrared probe, when the object appears in front of the probe, the infrared rays of the transmitting head are reflected to the receiving head, the probe outputs a pulse to count a counter, and the defects that the direction of a target cannot be judged, whether the target leaves cannot be confirmed, and the counting is easy to repeat.

Disclosure of Invention

The millimeter wave radar detection mainly applies Doppler effect and FMCW technology. The basic principle is that the transmitted wave is a frequency-modulated continuous wave, and the frequency of the frequency-modulated continuous wave changes along with the time according to the triangular wave rule. The frequency change rule of the received echo and the transmitted frequency is the same as a triangular wave rule, only a time difference exists, the target distance can be calculated by utilizing the small time difference, and radar point cloud containing distance, speed and direction information is obtained. The device has the advantages of high detection precision, all-weather work and no influence of weather. In view of this, the invention provides a method for counting the number of people passing in and out based on a millimeter wave radar.

The invention discloses a method for counting the number of people who enter and exit based on a millimeter wave radar, which comprises the following steps:

detecting a human body target by the millimeter wave radar, carrying out point cloud cluster tracking processing on the human body target, and outputting an output clustered target set T = { T =₁......,T_kWhere T is₁......,T_kK is the number of clustering targets;

adopting an extended Kalman tracking algorithm to perform clustering on the target set T = { T = { (T) }₁......,T_kEach target T in_iPredicting and updating, and acquiring the position information of the target in real time;

converting the position of the target Ti from a polar coordinate to a Cartesian coordinate system to obtain a tracking target set tracks = { TR = { (TR) }₀，TR₁，......TR_n-1In which TR_i=[T_id，X_i，Y_i，V_i]，X_i，Y_iIs the spatial coordinate, T, of the relative radar in the radar three-dimensional space_idIs track ID, V_iIs the speed;

continuously tracking the same track ID in the track list, judging the in-out state of the tracked target according to the position and the speed of the tracked target of the same track ID relative to the first crossing line, the second crossing line and the third crossing line, adding 1 to the number of the in-out persons of the target of which the tracked target is in an entried state, and subtracting 1 from the number of the in-out persons of the target of which the tracked target is in an exited state.

Further, the point cloud cluster tracking clustering process comprises the following steps:

clustering: selecting a first point from the point cloud set, and marking as P₀And is combined with P₀As centres of targets T to be clustered, i.e. T_kcenter=[R_center，V_center，θ_center]=[R₀，V₀，θ₀]Wherein R is_centerDistance of center point, V_centerVelocity at the center point, θ_centerAngle of center point, P₀And the point cloud P remains in the point cloud set_i(i = {1,2...., n-1} successive comparisons;

if conditions a, b are satisfied simultaneously:

a.|R_center*R_center+R_i*R_i-2R_center*R_i*cos(θ_center-θ_i)|<a first threshold value;

b.|V_center-V_i|<a second threshold value;

updating the central point of the target to be clustered,

T_center=[R_center,V_center,θ_center]=[(R_center+R_i)/2,(V_center+V_i)/2,(θ_center+θ_i)/2]，

wherein R is_iIs a target point cloud P _i Is a distance of_，V_iIs a target point cloud P _i Velocity of (e), theta_iIs a target point cloud P _i The angle of (d);

the point cloud P_iMarkingIs in a traversed state and is associated with a clustering target T_kLabel uniformity_；Target T_kContaining point clouds T_k={P_i,.......}；

iPlus 1, i.e. T to be updated_kcenterComparing with the next point in the point cloud set;

if the conditions a and b are not met, the central point of the target to be clustered is not required to be updated, and the central point of the target to be clustered is required to be updatediPlus 1, i.e. T_kcenterComparing with the next point in the point cloud set;

target T to be clustered_k，Comprises m point clouds, if m is larger than a preset minimum point cloud number threshold value P_numAnd the sum S of the SNR of all points_sumIf the energy is greater than the preset energy threshold value, T is judged_KIs a valid clustering target.

Further, when tracking the target T_iAt the k frame, X thereof_iUnknown region outside the first crossing line, and X at the k +1 th frame_iInside the first passing line while satisfying the velocity V_iIf the current state is unknown, the target is passing through the first crossing line, that is, the target enters the first region from the unknown region; its state becomes T_{i_state}=detected_incoming；

When tracking target T_iAt the k frame, X thereof_iOutside the second crossing line; and X at the k +1 th frame_iAt the inner side of the second crossing line, while satisfying the velocity V_iPositive and the current state is detected _ accompanying, indicating that the object is crossing the second crossing line 2, i.e. it enters the second area from the first area; its state becomes T_{i_state}=incoming；

When tracking target T_iX at the k frame_iOutside the third crossing line; and X at the k +1 th frame_iAt the inner side of the third passing line, while satisfying the velocity V_iPositive and the current state is incomming, which means that the target is passing through the third crossing line, i.e. it enters the third area from the second area; its state becomes T_{i_state}=entered；

Adding 1 to the total number of people in the room, and changing the state of the tracking target track ID into unknown againState, i.e. T_{i_state}=unknow。

Further, when tracking the target T_iX in the k frame_iInside the third crossing line; and X at the k +1 th frame_iOutside the third passing line while satisfying the velocity V_iNegative and the current state is unknown, it means that the object is passing through the third crossing line, i.e. it enters the second region from the third region; its state becomes T_{i_state}=detected_outgoing；

When tracking target T_iX at the k frame_iOn the inner side of the second crossing line; and X at the k +1 th frame_iOutside the second crossing line, while satisfying the velocity V_iNegative and the current state is detected _ outputting, indicating that the object is crossing the first crossing line, i.e. it enters the first area from the second area; its state becomes T_{i_state}=outgoing；

When tracking target T_iX at the k frame_iX at the k +1 th frame inside the first crossing line_iOutside the first passing line while satisfying the velocity V_iNegative and the current state is outgoing, it means that the object is crossing the first crossing line, i.e. it enters the unknown area from the first area; its state becomes T_{i_state}=exited；

Subtracting 1 from the total number of people in the room, and changing the state of the track ID of the tracking target into an unknown state, namely T_{i_state}=unknow。

The invention discloses a device for counting the number of people passing in and out based on a millimeter wave radar, which comprises a microprocessor, the millimeter wave radar, a battery and a wireless transmission module; the microprocessor is respectively connected with the millimeter wave radar and the wireless transmission module through serial ports; the device is installed in the middle of the top of an entrance and exit door, the normal direction of the millimeter wave radar is perpendicular to the ground, and the coordinate X direction of the millimeter wave radar is consistent with the entrance and exit direction.

Compared with the prior art, the invention has the following beneficial effects:

the invention is based on the millimeter wave radar, can distribute the only track ID to the pedestrian in the monitoring area, judge the position and speed direction of the same track ID on the adjacent frame at the same time; and four kinds of switching of the in-out state quantities are designed, so that the in-out number of the target is judged and counted, the stable counting of the number of the in-out people is ensured by switching of the four states, and the repeated counting of people staying at the doorway is avoided.

Drawings

FIG. 1 is a system framework of the present invention;

FIG. 2 is a diagram of the internal data processing process of the present invention;

FIG. 3 is a schematic view of a cross-over line arrangement according to the present invention;

FIG. 4 is a diagram of the in-out counting process of the present invention;

FIG. 5 is a diagram illustrating an in-out counting process according to an embodiment of the present invention.

In the figure, 1-the first crossing line, 2-the second crossing line, and 3-the third crossing line.

Detailed Description

The invention is further described with reference to the accompanying drawings, but the invention is not limited in any way, and any alterations or substitutions based on the teaching of the invention are within the scope of the invention.

As shown in fig. 1, the millimeter wave radar in-out people counting device is composed of a microprocessor, a millimeter wave radar, a battery and a wireless transmission module; the microprocessor is connected with the millimeter wave radar and the wireless transmission module through serial ports respectively. The device is installed in the middle of the top of an entrance and exit door, the normal direction of a radar is vertical to the ground, and the X direction of a radar coordinate is consistent with the entrance and exit direction.

Detecting a human body target by the millimeter wave radar, carrying out point cloud cluster tracking processing on the human body target, and outputting a cluster target set T = { T =₁......,T_kWhere T is₁......,T_kK is the number of clustering targets;

the clustering process in the point cloud cluster tracking is as follows:

clustering: selecting a first point from the point cloud set, and marking as P₀And is combined with P₀As centres of targets T to be clustered, i.e. T_kcenter=[R_center，V_center，θ_center]=[R₀，V₀，θ₀]，P₀And the point cloud P remains in the point cloud set_i(i = {1,2...., n-1} successive comparisons;

if conditions a, b are satisfied simultaneously:

a.|R_center*R_center+R_i*R_i-2R_center*R_i*cos(θ_center-θ_i)|<0.5；

b.|V_center-V_i|<1；

updating the central point of the target to be clustered:

T_center=[R_center,V_center,θ_center]=[(R_center+R_i)/2,(V_center+V_i)/2,(θ_center+θ_i)/2];

the point cloud P_iMarked as traversed state and associated with clustering target T_kLabel uniformity_；Target T_kContaining point clouds T_k={P_i,.......}；

iPlus 1, i.e. T to be updated_kcenterComparing with the next point in the point cloud set;

if the conditions a and b are not met, the central point of the target to be clustered is not required to be updated, and the central point of the target to be clustered is required to be updatediPlus 1, i.e. T_kcenterComparing with the next point in the point cloud set;

target T to be clustered_k，Comprises m point clouds, if m is larger than a preset minimum point cloud number threshold value P_numAnd the sum S of the SNR of all points_sumIf the energy is greater than the preset energy threshold value, T is judged_KIs a valid clustering target.

Adopting an extended Kalman tracking algorithm to perform clustering on the target set T = { T = { (T) }_1,......,T_kEach target T in_iPredicting and updating, and acquiring the position information of the target in real time; the distance and angle of information contained by each target, i.e. T_i={T_id，R_i，θ_i，V_i}；

Mixing Ti = { T_id，R_i,θ_i,V_iConverting from polar coordinates to Cartesian seatsIn the standard system, a tracking target set tracks = { TR = is obtained₀，TR₁，......,TR_n-1In which TR_i=[T_id，X_i，Y_i，V_i]，X_i，Y_iIs the spatial coordinate, T, of the relative radar in the radar three-dimensional space_idIs track ID, V_iIs the speed;

continuously tracking the same track ID in the track list, wherein the state of the track ID is switched after passing through a plurality of areas preset by a counting algorithm;

as shown in fig. 3, three crossing lines are provided, in which the crossing line 1 is a first crossing line, the crossing line 2 is a second crossing line, the radar is disposed directly above the crossing line 2, the crossing line 3 is a third crossing line, an area outside the first crossing line (i.e., in the direction toward the outside of the door) is an unknown area, a first area (i.e., area 1) is between the first crossing line and the second crossing line, a second area (i.e., area 2) is between the second crossing line and the third crossing line, and an area inside the third crossing line (i.e., in the direction toward the inside of the door) is a third area (i.e., area 3).

The method comprises the following steps:

when tracking target T_iAt the k frame, X thereof_iOn the outside of the gate (also called unknown region) of the crossing line 1; and at the k +1 th frame, X thereof_iOn the inner side of the door of the crossing line 1, while satisfying the velocity V_iPositive and the current state is unknown, it means that the target is crossing the crossing line 1, i.e. it enters the region 1 from an unknown region; its state becomes T_{i_state}=detected_incoming;

When tracking target T_iAt the k frame, X thereof_iOn the outside of the door of the crossing line 2; and at the k +1 th frame, X thereof_iOn the inner side of the door of the crossing line 2, while satisfying the velocity V_iPositive and the current state is detected _ accompanying, indicating that the object is crossing the crossing line 2, i.e. it is entering area 2 from area 1; its state becomes T_{i_state}=incoming;

When tracking target T_iAt the k frame, X thereof_iOn the outside of the door of the crossing line 3; and at the k +1 th frame, X thereof_iOn the inner side of the door across the thread 3,simultaneously satisfy the speed V_iPositive and the current state is incomming, indicating that the target is crossing cross-over line 3, i.e. it is entering area 3 from area 2; its state becomes T_{i_state}=entered;

Adding 1 to the total number of people in the room, changing the ID state to unknown state again after adding the total number of people in the room, T_{i_state}=unknow；

An outgoing process:

when tracking target T_iAt the k frame, X thereof_iInside the door of the crossing line 3; and at the k +1 th frame, X thereof_iOn the outside of the door through the line 3 while satisfying the velocity V_iNegative and the current state is unknown, indicating that the object is crossing the crossing line 3, i.e. it enters the region 2 from an unknown region (which now coincides with the region 3); its state becomes T_{i_state}=detected_outgoing;

When tracking target T_iAt the k frame, X thereof_iOn the inside of the door of the crossing line 2; and at the k +1 th frame, X thereof_iOn the outside of the door of the crossing line 2, while satisfying the velocity V_iNegative and the current state is detected _ outputting, this means that the target is crossing the crossing line 1, i.e. it is entering area 1 from area 2; its state becomes T_{i_state}=outgoing;

When tracking target T_iAt the k frame, X thereof_iOn the inside of the door of the crossing line 1; and at the k +1 th frame, X thereof_iOn the outside of the door of the crossing line 1, while satisfying the velocity V_iNegative and the current state is outgoing, it means that the object is crossing the crossing line 1, i.e. it is going from area 1 into unknown area; its state becomes T_{i_state}=exited;

The total number of people in the room is reduced by 1, and the state of the ID is changed into an unknown state again, T_{i_state}=unknow。

Examples

When the track ID is T_iThe pedestrian of (1) in the k-th frame, X thereof_iOn the outside of the door (X) of the crossing line 1_i<-1.5 m); and at the k +1 th frame, X thereof_iOn the inside of the door (X) of the crossing line 1_iNot less than-1.5 m) while satisfying the velocity V_iIs positive and the current state is unknow, this indicates that the target is crossing the crossing line 1.

In one embodiment, as shown in fig. 1, the millimeter wave radar in-out people counting device is composed of a microprocessor, a millimeter wave radar, a battery and a wireless module; the microprocessor is connected with the millimeter wave radar and the wireless transmission module through serial ports respectively. The device is arranged in the center of the top of an in-out door, the normal direction of a radar is vertical to the ground, and the X direction of a radar coordinate is consistent with the in-out direction.

Referring to fig. 2, a millimeter wave radar within a device transmits a millimeter-wave band radio frequency signal to a monitored area through its multiple-transmit multiple-receive (MIMO) antenna while receiving an echo signal scattered via a reflection point within the monitored area; the echo signal is down-converted into a beat intermediate frequency signal si (T) after being mixed with the transmitting signal, si (T) is subjected to ADC (analog to digital converter) sampling, distance dimension FFT (fast Fourier transform), speed estimation, CFAR (computational fluid dynamics) detection and horizontal and vertical angle estimation to obtain a dynamic moving pedestrian point cloud set in the current scene, and the point cloud is subjected to group tracking processing to obtain a tracking target set tracks = { T = (T) }₀，T₁，......T_n-1In which T is_i=[T_id,X_i,Y_i,V_i]，X_i,Y_iCan be regarded as the space coordinate, T, of the relative radar in the radar three-dimensional space_idIs track ID, V_iIs the velocity.

When each pedestrian passes through the monitoring area of the radar, a track with a unique ID can be formed; and judging the passing-in and passing-out state of the pedestrian by judging the position and the speed direction of the flight path. The method comprises the following specific steps:

suppose the position of the crossing line 1 is x₁=1.5m, the position of the crossing line 2 is x₂=0m, the position of the crossing line 3 is x₃=1.5 m; the speed direction that the radar monitored the pedestrian and got into is positive, and the speed direction that the radar monitored the pedestrian and got out is negative.

Referring to fig. 4, the entry counting process is as follows:

when the track ID is T_iThe pedestrian of (1) in the k-th frame, X thereof_iOn the outside of the door (X) of the crossing line 1_i<-1.5 m); and at the k +1 th frame, X thereof_iOn the inside of the door (X) of the crossing line 1_iNot less than-1.5 m) while satisfying the velocity V_iPositive and the current state is unknown, it means that the target is crossing the crossing line 1, i.e. it enters the region 1 from an unknown region; its state becomes T_{i_state}=detected_incoming；

When tracking target T_iAt the k frame, X thereof_iOn the outside of the door (X) of the crossing line 2_i<0 m); and at the k +1 th frame, X thereof_iOn the inside of the door (X) of the crossing line 2_iNot less than 0m) while satisfying the velocity V_iPositive and the current state is detected _ accompanying, indicating that the object is crossing the crossing line 2, i.e. it is entering area 2 from area 1; its state becomes T_{i_state}=incoming；

When tracking target T_iAt the k frame, X thereof_iOn the outside of the door (X) of the crossing line 3_i<1.5 m); and at the k +1 th frame, X thereof_iOn the inside of the door (X) across the line 3_iNot less than 1.5m) while satisfying the velocity V_iPositive and the current state is incomming, indicating that the target is crossing cross-over line 3, i.e. it is entering area 3 from area 2; its state becomes T_{i_state}=entered；

At this time, the total number of people in the room is increased by 1, and after the total number of people is increased, the state of the ID is changed into an unknown state again, T_{i_state}=unknow。

Other situations in the process of entering a person, such as a person crossing back and forth between a first area and a second area, will always keep its status in the last status (for example, a person is in the first area to the second area, the status changes to incoming, the target returns from the second area to the first area, the status does not change, or incoming is kept, so that the status switching from the second area to the third area is not affected).

Referring to fig. 5, the exit count process is as follows:

when tracking target T_iAt the k frame, X thereof_iOn the inside of the door (X) across the line 3_iMore than or equal to 1.5 m); and at the k +1 th frame, X thereof_iOn the outside of the door (X) of the crossing line 3_i<1.5m) while satisfying the velocity V_iNegative and the current state is unknown, this means that the target is passing through the crossing line 3, i.e. it is in an unknown region (in this case, the unknown region is overlapped with the region 3)And) into zone 2; its state becomes T_{i_state}=detected_outgoing;

When tracking target T_iAt the k frame, X thereof_iOn the inside of the door (X) of the crossing line 2_iMore than or equal to 0 m); and at the k +1 th frame, X thereof_iOn the outside of the door (X) of the crossing line 2_i<0m) while satisfying the velocity V_iNegative and the current state is detected _ outputting, this means that the object is crossing the crossing line 1, i.e. it enters the area 1 from the area 2; its state becomes T_{i_state}=outgoing;

When tracking target T_iAt the k frame, X thereof_iOn the inside of the door (X) of the crossing line 1_iMore than or equal to-1.5 m); and in the (k + 1) th frame (X)_i<-1.5m), X thereof_iOn the outside of the door of the crossing line 1, while satisfying the velocity V_iNegative and the current state is outgoing, it means that the object is crossing the crossing line 1, i.e. it is going from area 1 into unknown area; its state becomes T_{i_state}=exited。

The total number of people in the room is reduced by 1, and the state of the ID is changed into an unknown state again, T_{i_state}=unknow。

Other situations in the process of leaving a person, such as a person crossing back and forth between the second area and the first area, will always keep its status in the last state (e.g. a person in the second area goes to the first area, the status changes to outgoing, the target returns from the first area to the second area, the status does not change, or remains outgoing, and therefore does not affect its status switch from the first area to the unknown area (leaving)).

Compared with the prior art, the invention has the following beneficial effects:

the invention is based on the millimeter wave radar, can distribute the only track ID to the pedestrian in the monitoring area, judge the position and speed direction of the same track ID on the adjacent frame at the same time; and four kinds of switching of the in-out state quantities are designed, so that the in-out number of the target is judged and counted, the stable counting of the number of the in-out people is ensured by switching of the four states, and the repeated counting of people staying at the doorway is avoided.

The word "preferred" is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as "preferred" is not necessarily to be construed as advantageous over other aspects or designs. Rather, use of the word "preferred" is intended to present concepts in a concrete fashion. The term "or" as used in this application is intended to mean an inclusive "or" rather than an exclusive "or". That is, unless specified otherwise or clear from context, "X employs A or B" is intended to include either of the permutations as a matter of course. That is, if X employs A; b is used as X; or X employs both A and B, then "X employs A or B" is satisfied in any of the foregoing examples.

Also, although the disclosure has been shown and described with respect to one or an implementation, equivalent alterations and modifications will occur to others skilled in the art based upon a reading and understanding of this specification and the annexed drawings. The present disclosure includes all such modifications and alterations, and is limited only by the scope of the appended claims. In particular regard to the various functions performed by the above described components (e.g., elements, etc.), the terms used to describe such components are intended to correspond, unless otherwise indicated, to any component which performs the specified function of the described component (e.g., that is functionally equivalent), even though not structurally equivalent to the disclosed structure which performs the function in the herein illustrated exemplary implementations of the disclosure. In addition, while a particular feature of the disclosure may have been disclosed with respect to only one of several implementations, such feature may be combined with one or other features of the other implementations as may be desired and advantageous for a given or particular application. Furthermore, to the extent that the terms "includes," has, "" contains, "or variants thereof are used in either the detailed description or the claims, such terms are intended to be inclusive in a manner similar to the term" comprising.

Each functional unit in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or a plurality of or more than one unit are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium. The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Each apparatus or system described above may execute the storage method in the corresponding method embodiment.

In summary, the above-mentioned embodiment is an implementation manner of the present invention, but the implementation manner of the present invention is not limited by the above-mentioned embodiment, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be regarded as equivalent replacements within the protection scope of the present invention.

Claims (4)

1. A method for counting the number of people passing in and out based on a millimeter wave radar is characterized by comprising the following steps:

detecting a human body target by the millimeter wave radar, carrying out point cloud cluster tracking processing on the human body target, and outputting a cluster target set T ═ T₁......,T_kWhere T is₁......,T_kK is the number of clustering targets;

adopting an extended Kalman tracking algorithm to cluster the target set T ═ { T ═ T₁......,T_kEach target T in_iPredicting and updating, and acquiring the position information of the target in real time;

converting the position of the target Ti from a polar coordinate to a Cartesian coordinate system to obtain a tracking target set tracks ═ { TR ═ TR { (TR) }₀，TR₁，......TR_n-1In which TR_i＝[T_id，X_i，Y_i，V_i]，X_i，Y_iIs the spatial coordinate, T, of the relative radar in the radar three-dimensional space_idIs track ID, V_iIs the speed;

continuously tracking the same track ID in the track list, judging the in-out state of the tracked target according to the position and the speed of the tracked target of the same track ID relative to the first crossing line, the second crossing line and the third crossing line, adding 1 to the number of the in-out persons of the target of which the tracked target is in an entried state, and subtracting 1 from the number of the in-out persons of the target of which the tracked target is in an exited state;

wherein when tracking the target T_iAt the k frame, X thereof_iUnknown region outside the first crossing line, and X at the k +1 th frame_iInside the first passing line while satisfying the velocity V_iIf the current state is unknown, the target is passing through the first crossing line, that is, the target enters the first region from the unknown region; its state becomes T_{i_state}＝detected_incoming；

When tracking target T_iAt the k frame, X thereof_iOutside the second crossing line; and X at the k +1 th frame_iAt the inner side of the second crossing line, while satisfying the velocity V_iPositive and the current state is detected _ accompanying, indicating that the object is crossing the second crossing line 2, i.e. it enters the second area from the first area; its state becomes T_{i_state}＝incoming；

When tracking target T_iX at the k frame_iOutside the third crossing line; and X at the k +1 th frame_iAt the inner side of the third passing line, while satisfying the velocity V_iPositive and the current state is incomming, which means that the target is passing through the third crossing line, i.e. it enters the third area from the second area; its state becomes T_{i_state}＝entered；

Adding 1 to the total number of people in the room, and changing the state of the tracking target track ID into an unknown state, namely T_{i_state}＝unknow。

2. The millimeter wave radar-based people in and out counting method according to claim 1, wherein the clustering process in the point cloud cluster tracking is as follows:

clustering: selecting a first point from the point cloud set, and marking as P₀And is combined with P₀As centres of targets T to be clustered, i.e. T_kcenter＝[R_center，V_center，θ_center]＝[R₀，V₀，θ₀]Wherein R is_centerDistance of center point, V_centerVelocity at the center point, θ_centerAngle of center point, P₀With the remaining point cloudP_i(i ═ 1,2...., n-1} successive comparisons;

if conditions a, b are satisfied simultaneously:

a.|R_center*R_center+R_i*R_i-2R_center*R_i*cos(θ_center-θ_i)|<a first threshold value;

b.|V_center-V_i|<a second threshold value;

updating the central point of the target to be clustered:

T_center＝[R_center,V_center,θ_center]＝[(R_center+R_i)/2,(V_center+V_i)/2,(θ_center+θ_i)/2]，

wherein R is_iIs a target point cloud P_iDistance of (V)_iIs a target point cloud P_iVelocity of (e), theta_iIs a target point cloud P_iThe angle of (d);

point cloud P_iMarked as traversed state and associated with clustering target T_kThe labels are consistent; target T_kContaining point clouds T_k＝{P_i,.......}；

i plus 1, i.e. T after updating_kcenterComparing with the next point in the point cloud set;

if the conditions a and b are not met, the central point of the target to be clustered is not required to be updated, i is added with 1, namely T_kcenterComparing with the next point in the point cloud set;

target T to be clustered_kM point clouds are included, if m is larger than a preset minimum point cloud number threshold value P_numAnd the sum S of the SNR of all points_sumIf the energy is greater than the preset energy threshold value, T is judged_KIs a valid clustering target.

3. The millimeter wave radar-based entry-exit people counting method according to claim 1,

when tracking target T_iX in the k frame_iInside the third crossing line; and X at the k +1 th frame_iOutside the third passing line while satisfying the velocity V_iNegative and the current state is unknown, it means that the object is passing through the third crossing line, i.e. it enters the second region from the third region; its state becomes T_{i_state}＝detected_outgoing；

When tracking target T_iX at the k frame_iOn the inner side of the second crossing line; and X at the k +1 th frame_iOutside the second crossing line, while satisfying the velocity V_iNegative and the current state is detected _ outputting, indicating that the object is crossing the first crossing line, i.e. it enters the first area from the second area; its state becomes T_{i_state}＝outgoing；

When tracking target T_iX at the k frame_iX at the k +1 th frame inside the first crossing line_iOutside the first crossing line while satisfying the velocity V_iNegative and the current state is outgoing, it means that the object is crossing the first crossing line, i.e. it enters the unknown area from the first area; its state becomes T_{i_state}＝exited；

Subtracting 1 from the total number of people in the room, and changing the state of the track ID of the tracking target into an unknown state, namely T_{i_state}＝unknow。

4. A device for counting the number of people passing in and out based on a millimeter wave radar is characterized in that the device for counting the number of people passing in and out based on the millimeter wave radar is applied to the method for counting the number of people passing in and out based on the millimeter wave radar as claimed in any one of claims 1 to 3, and the device comprises a microprocessor, the millimeter wave radar, a battery and a wireless module; the microprocessor is respectively connected with the millimeter wave radar and the wireless transmission module through serial ports; the device is installed in the middle of the top of an entrance and exit door, the normal direction of the millimeter wave radar is perpendicular to the ground, and the coordinate X direction of the millimeter wave radar is consistent with the entrance and exit direction.

Images