CN116008936A - Human body track tracking detection method based on millimeter wave radar - Google Patents
Human body track tracking detection method based on millimeter wave radar Download PDFInfo
- Publication number
- CN116008936A CN116008936A CN202211513859.3A CN202211513859A CN116008936A CN 116008936 A CN116008936 A CN 116008936A CN 202211513859 A CN202211513859 A CN 202211513859A CN 116008936 A CN116008936 A CN 116008936A
- Authority
- CN
- China
- Prior art keywords
- human body
- track
- cost
- detection
- millimeter wave
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a human body track tracking detection method based on millimeter wave radar, and relates to the technical field of millimeter wave radar detection. The invention comprises the following steps: adopting a detection algorithm for the millimeter wave radar, and processing echo signals to obtain human body related information; predicting the human body position by adopting a Kalman filtering algorithm to obtain a predicted target position at the next moment; matching the detected human body position at the next moment with the predicted position at the next moment of the existing human body motion trail by adopting a global nearest neighbor algorithm; and updating the matched human body track information and correcting the currently estimated human body position. According to the human body tracking detection method, the human body target is obtained through millimeter wave radar detection, statistical analysis is carried out, related information is extracted, and the human body tracking detection precision is improved through human body track prejudgment and correction between actual tracks, so that the human body tracking detection method has the advantages of being strong in instantaneity, stable in tracking and strong in portability.
Description
Technical Field
The invention belongs to the technical field of millimeter wave radar detection, and particularly relates to a human body track tracking detection method based on millimeter wave radar, which can be widely applied to the fields of intelligent driving, security monitoring, intelligent home furnishing and the like.
Background
The millimeter wave radar is a radar for detecting the operation of the frequency domain of 30-300 GHZ, and has the advantages of a microwave radar and a photoelectric radar. The antenna has small antenna caliber and narrow beam; a large bandwidth; high Doppler frequency; good stealth resistance, etc.
Along with the development and application of millimeter wave radar detection technology in the fields of intelligent driving, security monitoring, intelligent home, military guidance and the like. In order to improve the use experience of consumers, better millimeter wave radar detection technology products are provided, and the millimeter wave radar detection technology has higher and higher requirements on human body detection.
A human body detection method and apparatus as in patent number CN202110134941.4, using an area detection algorithm to detect intermediate frequency signals and determine that a target enters a specified area, and using a multi-micro detection algorithm to detect intermediate frequency signals to determine whether a person is in the specified area, but cannot track and predict a human body trajectory.
Disclosure of Invention
The invention aims to provide a human body track tracking detection method based on millimeter wave radar, which is characterized in that related human body detection information is continuously acquired through a radar sensor, and a motion track is established according to the related human body detection information; and then, carrying out nearest neighbor matching on the human body detection result and the motion trail, and updating the motion trail information by adopting a Kalman filter, thereby solving the problem of insufficient accuracy of the existing millimeter wave human body detection trail tracking.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the invention relates to a human body track tracking detection method based on millimeter wave radar, which comprises the following steps:
step S1: adopting a detection algorithm for the millimeter wave radar, and processing echo signals to obtain human body related information;
step S2: predicting the human body position by adopting a Kalman filtering algorithm to obtain a predicted target position at the next moment;
step S3: matching the detected human body position at the next moment with the predicted position at the next moment of the existing human body motion trail by adopting a global nearest neighbor algorithm; if a plurality of targets exist in the scene, the targets are distributed to different human body motion tracks according to an algorithm;
step S4: updating the matched human body track information, correcting the current estimated human body position, updating the unmatched human body motion track and recording the unmatched human body motion track as an invisible state;
step S5: deleting the track of continuous multi-frames with the total invisible frame number exceeding the threshold value and maintaining the invisible state, and establishing a corresponding motion track for unmatched measurement in the step S3;
in step S3, the nearest neighbor algorithm is adopted to match the detection result with the tracking prediction track, and the specific flow is as follows:
step S31: judging the number N and the detection number M of tracks in the current frame, and adopting different solutions according to the combination of the number N and the detection number M;
step S32: generating a Cost matrix Cost [ N ] [ M ];
step S33: generating an edge matrix Padded_cost [ N+M ];
step S34: and judging the position of the Padded_cost [ rows ] [ cols ], and generating a motion trail.
As a preferable technical solution, in the step S1, the human body related information is position information, speed information and angle information of the target human body acquired by the millimeter wave radar.
In the step S2, the kalman filtering algorithm uses high frequency to collect data frames, and human body motion between frames can be regarded as uniform motion; let the human transverse and longitudinal coordinates at k moment be x (k) and y (k) respectively, and the speeds be v respectively x (k) And v y (k) The calculation formula at time k+1 is as follows:
state X at time k+1 k+1 The method comprises the following steps:
X k+1 ＝[x(k+1) v x (k+1) y(k+1) v y (k+1)] T ；
the state equation of the system is:
X k+1 ＝AX k ；
the state transition matrix is:
the position information of the target is acquired in the observation process, and the measurement equation is as follows:
Z k+1 ＝HX k+1 +V k+1 ；
v in k In order to measure the noise of the light,
the measurement at time k+1 is:
Z k+1 ＝[z x (k+1) z y (k+1)] T ＝[x(k+1) y(k+1)] T 。
as a preferred technical solution, in step S31, the combination condition of the number N of inner tracks and the number M of detection tracks is determined as follows:
if there is no trace and no detection (n=0 and m=0), then jumping out of the function;
if tracks exist but are not detected (N is not equal to 0 and M is not equal to 0), the sequence numbers of all effective tracks in the current frame are recorded and stored in an unmatched track array, and then the function is jumped out;
if there is detection but no track (n=0 and M is not equal to 0), recording all the detected sequence numbers in the current frame and storing them into unmatched detection array, then jumping out the function;
if there is a trace and there is a detection (N.noteq.0 and M.noteq.0), the Markov distance between the trace and the detection is calculated, and a Cost matrix Cost [ N ] [ M ] is generated.
As a preferred technical solution, in the step S32, a specific procedure for generating the Cost matrix Cost [ N ] [ M ] is as follows:
step S321: obtaining the horizontal and vertical coordinates of the central point of the ith track;
step S322: calculating the mahalanobis distance between the ith track and the jth track;
wherein, x, y is the horizontal and vertical coordinates of the track center point, and x1, y1 is the horizontal and vertical coordinates of the detection center point;
step S323: generating a matrix Cost (i, j) =euclidean;
step S324: traversing the Cost matrix [ N ] [ M ], and finding the maximum value max of the matrix.
As a preferable technical solution, in the step S33, a edge matrix padded_cost [ n+m ]; wherein, the elements on the diagonal of Padded_Cost [ N ] [ N ] and Padded_Cost [ M ] [ M ] are equal to 30, and the rest elements are max.
As a preferred embodiment, in the step S34,
traversing each column of the edge matrix if the Padded_Cost [ N+M ] edge matrix N > M, and if the Padded_Cost [ rows ] [ cols ] is the minimum element min of the cols column, then Padded_Cost [ rows, cols+1:M+N ] =31;
otherwise, traversing each row of the edge matrix, if Padded_cost [ rows ] [ cols ] is the smallest element min of the rows column, padded_cost [ rows+1:N+M, cols ] =31.
As a preferable technical scheme, the position judgment method of the Padded_cost [ rows ] [ cols ] is as follows:
when row is less than N & cols is less than M, detecting cols and matching with the row track, adding 1 to the number of matched tracks, updating the motion track, judging whether the number of tracks exceeds a threshold value, and deleting the initial track if the number of tracks exceeds the threshold value;
when rows is less than N & M and less than or equal to col and less than N+M, the rows track is an unmatched track, 1 is added to the number of unmatched tracks, the tracks are marked as invisible states, and the tracks are updated;
when N is less than or equal to rows and less than N+M and cols is less than M, the cols bar is detected as unmatched detection, and 1 is added to the unmatched detection number, so that a new motion track is created.
The invention has the following beneficial effects:
the human body target is obtained by utilizing millimeter wave radar detection, statistical analysis is performed, related information is extracted, and human body track tracking detection precision is improved by performing correction between human body track prejudgment and actual tracks, so that the method has the advantages of being strong in instantaneity, stable in tracking and strong in portability.
Of course, it is not necessary for any one product to practice the invention to achieve all of the advantages set forth above at the same time.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a human body track tracking detection method based on millimeter wave radar;
FIG. 2 is a schematic diagram of an application of a method for tracking and detecting human body trajectories based on millimeter wave radar to an air conditioner;
FIG. 3 is a schematic diagram of the resulting edge matrix Padded_cost [ N+M ].
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the invention discloses a human body track tracking detection method based on millimeter wave radar, which comprises the following steps:
step S1: adopting a detection algorithm for the millimeter wave radar, and processing echo signals to obtain human body related information, wherein the human body related information is the position information, the speed information, the angle information and the like of a target human body obtained by the millimeter wave radar;
step S2: predicting the human body position by adopting a Kalman filtering algorithm to obtain a predicted target position at the next moment;
in the step S2, the Kalman filtering algorithm adopts high frequency to collect data frames, and human body motion between frames can be regarded as uniform motion; let the human transverse and longitudinal coordinates at k moment be x (k) and y (k) respectively, and the speeds be v respectively x (k) And v y (k) The calculation formula at time k+1 is as follows:
state X at time k+1 k+1 The method comprises the following steps:
X k+1 ＝[x(k+1)v x (k+1)y(k+1)v y (k+1)] T ；
the state equation of the system is:
X k+1 ＝AX k ；
the state transition matrix is:
the position information of the target is acquired in the observation process, and the measurement equation is as follows:
Z k+1 ＝HX k+1 +V k+1 ；
v in k In order to measure the noise of the light,
the measurement at time k+1 is:
Z k+1 ＝[z x (k+1)z y (k+1)] T ＝[x(k+1)y(k+1)] T 。
step S3: matching the detected human body position at the next moment with the predicted position at the next moment of the existing human body motion trail by adopting a global nearest neighbor algorithm; if a plurality of targets exist in the scene, the targets are distributed to different human body motion tracks according to an algorithm;
in step S3, the nearest neighbor algorithm is adopted to match the detection result with the tracking prediction track, and the specific flow is as follows:
step S31: judging the number N and the detection number M of tracks in the current frame, and adopting different solutions according to the combination of the number N and the detection number M;
step S32: generating a Cost matrix Cost [ N ] [ M ];
step S33: generating an edge matrix Padded_cost [ N+M ];
step S34: and judging the position of the Padded_cost [ rows ] [ cols ], and generating a motion trail.
Step S4: updating the matched human body track information, correcting the current estimated human body position, updating the unmatched human body motion track and recording the unmatched human body motion track as an invisible state;
step S5: deleting the track of continuous multi-frames with the total invisible frame number exceeding the threshold value and maintaining the invisible state, and establishing a corresponding motion track for unmatched measurement in the step S3;
in step S31, the combination of the number N of inner tracks and the number M of detection tracks is determined as follows:
if there is no trace and no detection (n=0 and m=0), then jumping out of the function;
if tracks exist but are not detected (N is not equal to 0 and M is not equal to 0), the sequence numbers of all effective tracks in the current frame are recorded and stored in an unmatched track array, and then the function is jumped out;
if there is detection but no track (n=0 and M is not equal to 0), recording all the detected sequence numbers in the current frame and storing them into unmatched detection array, then jumping out the function;
if there is a trace and there is a detection (N.noteq.0 and M.noteq.0), the Markov distance between the trace and the detection is calculated, and a Cost matrix Cost [ N ] [ M ] is generated.
In step S32, the specific flow of generating the Cost matrix Cost [ N ] [ M ] is as follows:
step S321: obtaining the horizontal and vertical coordinates of the central point of the ith track;
step S322: calculating the mahalanobis distance between the ith track and the jth track;
wherein, x, y is the horizontal and vertical coordinates of the track center point, and x1, y1 is the horizontal and vertical coordinates of the detection center point;
step S323: generating a matrix Cost (i, j) =euclidean;
step S324: traversing the Cost matrix [ N ] [ M ], and finding the maximum value max of the matrix.
Referring to FIG. 3, in step S33, a edge matrix Padded_cost [ N+M ] is generated; wherein, the elements on the diagonal of Padded_Cost [ N ] [ N ] and Padded_Cost [ M ] [ M ] are equal to 30, and the rest elements are max.
In step S34, if the packed_cost [ n+m ] edge matrix N > M, traversing each column of the edge matrix, and if the packed_cost [ rows ] [ cols ] is the minimum element min of the cols column, then the packed_cost [ rows, cols+1:m+n ] =31;
otherwise, traversing each row of the edge matrix, if Padded_cost [ rows ] [ cols ] is the smallest element min of the rows column, padded_cost [ rows+1:N+M, cols ] =31.
The position of the Padded_cost [ rows ] [ cols ] is determined as follows:
when row is less than N & cols is less than M, (namely min falls in a first area Cost [ N ] [ M ]), the colth strip is detected to be matched with the row's track, the number of matched tracks is added with 1, the motion track is updated, whether the number of tracks exceeds a threshold value is judged, and if the number of tracks exceeds the threshold value, the initial track is deleted;
when rows is less than N & M and less than or equal to col and less than N+M (namely min falls in a second area of the bordered matrix), the rows track is an unmatched track, the number of unmatched tracks is increased by 1, the tracks are marked as invisible states, and the tracks are updated;
when N is less than or equal to rows and less than N+M and cols is less than M (namely min falls in the area of the third bordered matrix), the cols bar is detected as unmatched detection, and the unmatched detection number is added with 1, so that a new motion track is created.
Example 1
As shown in fig. 1, in the human body detection stage, human body detection results in a scene are continuously acquired through a sensor, and human body related information is obtained through statistics. In the human body tracking stage, a Kalman filter is adopted for predicting the human body position in the motion trail, a nearest neighbor algorithm is used for continuously matching the human body position detected at the next moment, the human body motion trail information is selected to be updated or a new motion trail is created according to the matching result, and the motion trail exceeding a threshold value is deleted, so that the human body trail tracking detection method suitable for the millimeter wave radar is realized, and the specific flow is as follows:
and continuously acquiring target human body position information, speed information and angle information by the statistic human body reference information through the millimeter wave radar.
The human body position prediction adopts a Kalman filtering algorithm to predict the human body position, and the human body predicted position at the next moment is obtained. The high frequency is adopted to collect the data frames, and the human body movement between the frames can be regarded as uniform movement. The transverse and longitudinal coordinates of the human body at the moment k are respectively x (k) and y (k), and the speeds are respectively v x (k) And v y (k) Then time k+1:
state X at time k+1 k+1 The method comprises the following steps:
X k+1 ＝[x(k+1)v x (k+1)y(k+1)v y (k+1)] T
the system state equation at this time is:
X k+1 ＝AX k
the state transition matrix is:
the position information of the target is obtained in the observation process, so that the measurement equation is as follows:
Z k+1 ＝HX k+1 +V k+1
wherein V is k In order to measure the noise of the light,
the measurement at time k+1 is:
Z k+1 ＝[z x (k+1)z y (k+1)] T ＝[x(k+1)y(k+1)] T
the human body position tracking matching adopts a nearest neighbor algorithm to match the detection result with the tracking prediction track. The specific flow is as follows:
judging the number N of tracks in the current frame and the number M of detection tracks, and adopting different solutions according to the combination of the number N of tracks and the number M of detection tracks.
If there is no trace and no detection (n=0 and m=0), jumping out of the function;
if there is track but no detection (N +.0 and m=0), the sequence numbers of all valid tracks in the current frame are recorded and stored in the unmatched track array, and then the function is jumped out.
If there is a detected but no track (n=0 and m+.0), all detected sequence numbers in the current frame are recorded and stored in the unmatched detection array, and then the function is skipped.
If there is a trace and there is a detection (N.noteq.0 and M.noteq.0), the Markov distance between the trace and the detection is calculated, thereby generating a Cost matrix Cost [ N ] [ M ].
1. Obtaining the horizontal and vertical coordinates of the central point of the ith track;
2. calculating the mahalanobis distance between the ith track and the jth track;
where x, y are the horizontal and vertical coordinates of the center point of the track, and x1, y1 are the horizontal and vertical coordinates of the center point of the detection.
3. Generating a matrix Cost (i, j) =euclidean;
4. traversing the Cost matrix [ N ] [ M ], and finding the maximum value max of the matrix.
Generating an edge matrix Padded_cost [ N+M ]; wherein the elements on the diagonals of Padded_Cost [ N ] [ N ] and Padded_Cost [ M ] [ M ] are equal to 30, and the remaining elements are max.
Matching core algorithm
If the packed_cost [ n+m ] edge matrix N > M, traversing each column of the edge matrix, and if the packed_cost [ rows ] [ cols ] is the smallest element min of the cols column, then the packed_cost [ rows, cols+1:m+n ] =31; otherwise, traversing each row of the edge matrix, if padded_cost [ rows ] [ cols ] is the smallest element min of the rows column, then padded_cost [ rows+1:N+M, cols ] = 31.
Judging the position of the Padded_cost [ rows ] [ cols ]:
when row is less than N & cols is less than M (namely min falls in a first area Cost [ N ] [ M ]), the colth strip is detected to be matched with the row track, the number of matched tracks is added with 1, the motion track is updated, whether the number of tracks exceeds a threshold value is judged, and if the number of tracks exceeds the threshold value, the initial track is deleted.
When rows is less than N & M and less than or equal to col is less than N+M (namely min falls in the area of the bordered matrix II), the rows track is an unmatched track, the number of unmatched tracks is added with 1, the tracks are marked as invisible states, and the tracks are updated.
When N is less than or equal to rows and less than N+M and cols is less than M (namely min falls in the area of the third bordered matrix), the cols bar is detected as unmatched detection, and the unmatched detection number is added with 1, so that a new motion track is created.
Example two
As shown in fig. 2, the method can be applied to an air conditioner on the basis of the first embodiment, the position information of the person can be obtained in real time according to the track management information, strong air is sent from the air conditioner when the target is far from the air conditioner, soft air is sent from the air conditioner when the target is near to the air conditioner, and when the position of the human body changes, the air supply angle of the air conditioner changes along with the change of the position of the human body.
Therefore, the intelligent control of the air supply of the air conditioner is realized, the wind is driven by people, the distance is thrown and the distance is flexible, and the intelligent control device can be widely applied to the fields of intelligent driving, security monitoring, intelligent home furnishing and the like.
It should be noted that, in the above system embodiment, each unit included is only divided according to the functional logic, but not limited to the above division, so long as the corresponding function can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present invention.
In addition, those skilled in the art will appreciate that all or part of the steps in implementing the methods of the embodiments described above may be implemented by a program to instruct related hardware, and the corresponding program may be stored in a computer readable storage medium.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (8)
1. The human body track tracking and detecting method based on the millimeter wave radar is characterized by comprising the following steps of:
step S1: adopting a detection algorithm for the millimeter wave radar, and processing echo signals to obtain human body related information;
step S2: predicting the human body position by adopting a Kalman filtering algorithm to obtain a predicted target position at the next moment;
step S3: matching the detected human body position at the next moment with the predicted position at the next moment of the existing human body motion trail by adopting a global nearest neighbor algorithm; if a plurality of targets exist in the scene, the targets are distributed to different human body motion tracks according to an algorithm;
step S4: updating the matched human body track information, correcting the current estimated human body position, updating the unmatched human body motion track and recording the unmatched human body motion track as an invisible state;
step S5: deleting the track of continuous multi-frames with the total invisible frame number exceeding the threshold value and maintaining the invisible state, and establishing a corresponding motion track for unmatched measurement in the step S3;
in step S3, the nearest neighbor algorithm is adopted to match the detection result with the tracking prediction track, and the specific flow is as follows:
step S31: judging the number N and the detection number M of tracks in the current frame, and adopting different solutions according to the combination of the number N and the detection number M;
step S32: generating a Cost matrix Cost [ N ] [ M ];
step S33: generating an edge matrix Padded_cost [ N+M ];
step S34: and judging the position of the Padded_cost [ rows ] [ cols ], and generating a motion trail.
2. The human body track tracking detection method based on millimeter wave radar according to claim 1, wherein in the step S1, the human body related information is position information, speed information and angle information of a target human body acquired by the millimeter wave radar.
3. The human body track tracking detection method based on millimeter wave radar according to claim 1, wherein in the step S2, a kalman filtering algorithm adopts high frequency to perform data frame acquisition, and human body motion between frames can be regarded as uniform motion; let the human transverse and longitudinal coordinates at k moment be x (k) and y (k) respectively, and the speeds be v respectively x (k) And v y (k) The calculation formula at time k+1 is as follows:
state X at time k+1 k+1 The method comprises the following steps:
X k+1 ＝[x(k+1) v x (k+1) y(k+1) v y (k+1)] T ；
the state equation of the system is:
X k+1 ＝AX k ；
the state transition matrix is:
the position information of the target is acquired in the observation process, and the measurement equation is as follows:
Z k+1 ＝HX k+1 +V k+1 ；
v in k In order to measure the noise of the light,
the measurement at time k+1 is:
Z k+1 ＝[z x (k+1) z y (k+1)] T ＝[x(k+1) y(k+1)] T 。
4. the human body track tracking detection method based on millimeter wave radar according to claim 1, wherein in the step S31, the combination condition of the number N of inner tracks and the number M of detection is determined as follows:
if there is no trace and no detection (n=0 and m=0), then jumping out of the function;
if tracks exist but are not detected (N is not equal to 0 and M is not equal to 0), the sequence numbers of all effective tracks in the current frame are recorded and stored in an unmatched track array, and then the function is jumped out;
if there is detection but no track (n=0 and M is not equal to 0), recording all the detected sequence numbers in the current frame and storing them into unmatched detection array, then jumping out the function;
if there is a trace and there is a detection (N.noteq.0 and M.noteq.0), the Markov distance between the trace and the detection is calculated, and a Cost matrix Cost [ N ] [ M ] is generated.
5. The human body track tracking detection method based on millimeter wave radar according to claim 1, wherein in the step S32, the specific flow of generating the Cost matrix Cost [ N ] [ M ] is as follows:
step S321: obtaining the horizontal and vertical coordinates of the central point of the ith track;
step S322: calculating the mahalanobis distance between the ith track and the jth track;
wherein, x, y is the horizontal and vertical coordinates of the track center point, and x1, y1 is the horizontal and vertical coordinates of the detection center point;
step S323: generating a matrix Cost (i, j) =euclidean;
step S324: traversing the Cost matrix [ N ] [ M ], and finding the maximum value max of the matrix.
6. The human body track tracking detection method based on millimeter wave radar according to claim 1, wherein in the step S33, a edge matrix padded_cost [ n+m ] is generated; wherein, the elements on the diagonal of Padded_Cost [ N ] [ N ] and Padded_Cost [ M ] [ M ] are equal to 30, and the rest elements are max.
7. The method for human body trace detection based on millimeter wave radar according to claim 1, wherein in step S34,
traversing each column of the edge matrix if the Padded_Cost [ N+M ] edge matrix N > M, and if the Padded_Cost [ rows ] [ cols ] is the minimum element min of the cols column, then Padded_Cost [ rows, cols+1:M+N ] =31;
otherwise, traversing each row of the edge matrix, if Padded_cost [ rows ] [ cols ] is the smallest element min of the rows column, padded_cost [ rows+1:N+M, cols ] =31.
8. The human body track tracking detection method based on millimeter wave radar according to claim 7, wherein the position judgment mode of the padded_cost [ rows ] [ cols ] is as follows:
when row is less than N & cols is less than M, detecting cols and matching with the row track, adding 1 to the number of matched tracks, updating the motion track, judging whether the number of tracks exceeds a threshold value, and deleting the initial track if the number of tracks exceeds the threshold value;
when rows is less than N & M and less than or equal to col and less than N+M, the rows track is an unmatched track, 1 is added to the number of unmatched tracks, the tracks are marked as invisible states, and the tracks are updated;
when N is less than or equal to rows and less than N+M and cols is less than M, the cols bar is detected as unmatched detection, and 1 is added to the unmatched detection number, so that a new motion track is created.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211513859.3A CN116008936A (en) | 2022-11-29 | 2022-11-29 | Human body track tracking detection method based on millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211513859.3A CN116008936A (en) | 2022-11-29 | 2022-11-29 | Human body track tracking detection method based on millimeter wave radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116008936A true CN116008936A (en) | 2023-04-25 |
Family
ID=86034350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211513859.3A Pending CN116008936A (en) | 2022-11-29 | 2022-11-29 | Human body track tracking detection method based on millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116008936A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117524413A (en) * | 2024-01-05 | 2024-02-06 | 亿慧云智能科技（深圳）股份有限公司 | Motion protection method, device, equipment and medium based on millimeter wave radar |
-
2022
- 2022-11-29 CN CN202211513859.3A patent/CN116008936A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117524413A (en) * | 2024-01-05 | 2024-02-06 | 亿慧云智能科技（深圳）股份有限公司 | Motion protection method, device, equipment and medium based on millimeter wave radar |
| CN117524413B (en) * | 2024-01-05 | 2024-03-19 | 亿慧云智能科技（深圳）股份有限公司 | Motion protection method, device, equipment and medium based on millimeter wave radar |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102881024B (en) | Tracking-learning-detection (TLD)-based video object tracking method | |
| CN105717505B (en) | The data correlation method of multiple target tracking is carried out using Sensor Network | |
| CN113506317B (en) | Multi-target tracking method based on Mask R-CNN and apparent feature fusion | |
| CN100531405C (en) | Target tracking method of sports video | |
| US20150009323A1 (en) | Multi-target tracking method for video surveillance | |
| CN108470353A (en) | A kind of method for tracking target, device and storage medium | |
| CN114299417A (en) | Multi-target tracking method based on radar-vision fusion | |
| CN108573496B (en) | Multi-target tracking method based on LSTM network and deep reinforcement learning | |
| CN110927712B (en) | Tracking method and device | |
| CN105631895A (en) | Temporal-spatial context video target tracking method combining particle filtering | |
| CN101770024A (en) | Multi-target tracking method | |
| CN106226751A (en) | Maneu-vering target detection based on DP TBD and tracking | |
| WO2013160688A1 (en) | Abnormal object track determination using a gaussian processes based variational bayes expectation maximisation | |
| CN104793200B (en) | Dynamic planning track-before-detect method based on iterative processing | |
| CN109239702B (en) | Airport low-altitude flying bird number statistical method based on target state set | |
| Abdelkader et al. | Integrated motion detection and tracking for visual surveillance | |
| CN116008936A (en) | Human body track tracking detection method based on millimeter wave radar | |
| CN116645396A (en) | Track determination method, track determination device, computer-readable storage medium and electronic device | |
| CN104407345B (en) | Improved dynamic planning method for weak target detection tracking | |
| CN108182693A (en) | A kind of multiple target tracking algorithm based on tracking segment confidence level and appearance study | |
| KR20070067095A (en) | Method for detecting and tracking pointlike targets, in an optronic surveillance system | |
| CN108983194B (en) | Target extraction and condensation method based on ground monitoring radar system | |
| CN105974402A (en) | Maneuvering target multi-frame track-before-detect method applied to pulse-Doppler radar | |
| CN115469303A (en) | Cognitive biological radar method and device for detecting human body posture and vital signs | |
| CN108828584B (en) | Multi-frequency target tracking-before-detection method based on track folding factor ambiguity resolution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |