CN117908003B - Space people counting method and system - Google Patents
Space people counting method and system Download PDFInfo
- Publication number
- CN117908003B CN117908003B CN202410309379.8A CN202410309379A CN117908003B CN 117908003 B CN117908003 B CN 117908003B CN 202410309379 A CN202410309379 A CN 202410309379A CN 117908003 B CN117908003 B CN 117908003B
- Authority
- CN
- China
- Prior art keywords
- target
- space
- obtaining
- distance
- people
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000003068 static effect Effects 0.000 claims abstract description 41
- 238000001514 detection method Methods 0.000 claims abstract description 38
- 238000001228 spectrum Methods 0.000 claims abstract description 37
- 238000009434 installation Methods 0.000 claims abstract description 33
- 238000007621 cluster analysis Methods 0.000 claims abstract description 13
- 238000000638 solvent extraction Methods 0.000 claims description 15
- 238000005259 measurement Methods 0.000 claims description 7
- 230000000241 respiratory effect Effects 0.000 claims description 7
- 230000007547 defect Effects 0.000 abstract description 4
- 238000005192 partition Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000012217 deletion Methods 0.000 description 3
- 230000037430 deletion Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000003313 weakening effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/50—Systems of measurement based on relative movement of target
- G01S13/52—Discriminating between fixed and moving objects or between objects moving at different speeds
- G01S13/56—Discriminating between fixed and moving objects or between objects moving at different speeds for presence detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/06—Systems determining position data of a target
- G01S13/42—Simultaneous measurement of distance and other co-ordinates
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Electromagnetism (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a space people counting method and a space people counting system, which relate to the technical field of target detection and comprise the following steps: obtaining a preconfigured people counting space, wherein the people counting space comprises a space region division result; detecting the statistical space of the number of people by a radar to obtain a distance-Doppler spectrum; according to the radar installation position, obtaining a target point cloud by carrying out angle estimation on a target unit detected by the distance-Doppler spectrum; performing cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions; matching the number of distribution areas according to the distribution positions of the plurality of clustering centers to obtain the number of dynamic targets; traversing the space region division result to perform inching feature detection to obtain the number of stationary targets; and obtaining a statistical result of the number of people according to the number of the dynamic targets and the number of the static targets. The technical problem that the number of people in a specific space cannot be accurately detected due to the defects of multipath, lack of associated point clouds of a static target and the like is solved.
Description
Technical Field
The invention relates to the technical field of target detection, in particular to a space people counting method and a space people counting system.
Background
The statistics of the number of people by the function of the radar for person detection is one of the main application directions of the radar. The scene that needs to be checked for the number of people, such as hotel room statistics, factory workshop staff number, etc., can be applied.
The current statistical scheme for acquiring the number of people in a specific space such as hotel rooms, factory workshop staff and the like by using a radar mainly comprises the steps of processing point cloud data detected by the radar and determining the number and the position of the people in the specific space by using a target tracking algorithm.
The prior art has the defects of multipath problem, lack of associated point cloud of a static target and the like, so that the number of people in a specific space cannot be accurately detected.
Disclosure of Invention
The application provides a space people counting method and a space people counting system, which are used for solving the technical problem that the number of people in a specific space cannot be accurately detected due to the defects of multipath problem, lack of associated point cloud of a static target and the like in the prior art.
In view of the above problems, the present application provides a method and a system for counting the number of people in space.
In a first aspect of the present application, there is provided a method of spatial people counting comprising: obtaining a preconfigured people counting space, wherein the people counting space comprises a space region division result; detecting the people counting space by using a radar to obtain a distance-Doppler spectrum; according to the radar installation position, obtaining a target point cloud by carrying out angle estimation on a target unit detected by the distance-Doppler spectrum; performing cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions; according to the distribution positions of the plurality of clustering centers, matching the distribution area number of the space area division result to obtain the dynamic target number; traversing the space region division result to perform inching feature detection to obtain the number of stationary targets; and obtaining a statistical result of the number of people according to the number of the dynamic targets and the number of the static targets.
In a second aspect of the present application, there is provided a spatial people counting system comprising: the space acquisition module is used for acquiring a pre-configured people counting space, wherein the people counting space comprises a space region division result; the radar detection module is used for detecting the people counting space through a radar to obtain a distance-Doppler spectrum; the point cloud obtaining module is used for obtaining a target point cloud by carrying out angle estimation on the target unit detected by the distance-Doppler spectrum according to the radar installation position; the point cloud clustering module is used for carrying out cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions; the dynamic target sorting module is used for matching the number of the distribution areas of the space area division result according to the distribution positions of the plurality of clustering centers to obtain the number of dynamic targets; the static target sorting module is used for traversing the space region division result to detect micro-motion characteristics and obtain the number of static targets; and the people counting module is used for obtaining a people counting result according to the dynamic target quantity and the static target quantity.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The application determines the quantity of stationary targets by performing regional division on the space to be detected and performing micro-motion detection; meanwhile, determining the target point cloud for weakening the influence of multipath in the people counting space by combining the distance-Doppler spectrum with angle estimation; further clustering the target point cloud to obtain the number of the region categories corresponding to the distribution positions of the plurality of clustering centers, and determining the number of dynamic targets; and obtaining a statistical result of the number of people according to the number of the dynamic targets and the number of the static targets. The detection of the static target is realized by utilizing the micro-motion feature detection, the dynamic target with higher accuracy is obtained by utilizing the combination of the distance-Doppler spectrum and the angle estimation and then clustering, and the technical effect of improving the statistical accuracy of the number of people in a specific space is achieved.
Drawings
FIG. 1 is a schematic flow chart of a method for counting the number of people in space;
FIG. 2 is a schematic diagram of a spatial partition in a spatial people counting method according to the present application;
FIG. 3 is a schematic flow chart of detecting jog features in a space population statistics method according to the present application;
Fig. 4 is a schematic structural diagram of a spatial people counting system according to the present application.
Reference numerals illustrate: the system comprises a space acquisition module 100, a radar detection module 200, a point cloud acquisition module 300, a point cloud clustering module 400, a dynamic target sorting module 500, a static target sorting module 600 and a people counting module 700.
Detailed Description
The application provides a space people counting method and a space people counting system, which are used for solving the technical problem that the number of people in a specific space cannot be accurately detected due to the defects of multipath problem, lack of associated point cloud of a stationary target and the like in the prior art. The detection of the static target is realized by utilizing the micro-motion feature detection, the dynamic target with higher accuracy is obtained by utilizing the combination of the distance-Doppler spectrum and the angle estimation and then clustering, and the technical effect of improving the statistical accuracy of the number of people in a specific space is achieved.
Example 1
As shown in fig. 1, the present application provides a space people counting method, comprising the steps of:
S10: obtaining a preconfigured people counting space, wherein the people counting space comprises a space region division result;
Specifically, the demographics space refers to a space range in which demographics are required, and is exemplified by: when the number of hotel rooms is counted, each room can be a counting space for the number of people. In order to more accurately count the number of people, the people counting space is divided into a plurality of space areas, and the space area division results are stored, preferably, the size of each area is based on the area capable of accommodating one person, and the user adaptively sets the area based on the principle.
Furthermore, the embodiment of the application provides an unlimited space region division mode:
further, obtaining a preconfigured demographic space, comprising:
Obtaining a radar installation position;
Obtaining space range basic information, wherein the space range basic information comprises a left boundary constraint distance, a right boundary constraint distance and a front boundary constraint distance;
configuring a people counting space based on the radar installation position according to the left boundary constraint distance, the right boundary constraint distance and the front boundary constraint distance;
and obtaining a circumferential partitioning angle threshold and a forward partitioning distance threshold, and partitioning the people counting space based on the radar installation position to obtain a space region partitioning result.
Specifically, the radar installation position refers to the installation position of a radar sensor for detecting the number of people, the space range basic information refers to constraint parameters of the number of people counting space, the constraint parameters are preset by a user according to the radar installation position, the constraint parameters specifically comprise left boundary constraint distances, the detection distance on the left side of the radar installation position, and the left boundary of the number of people counting space can be determined based on the constraint parameters; the method further comprises a right boundary constraint distance, namely a left detection distance of the radar installation position, on the basis of which the right boundary of the people counting space can be determined, and a front boundary constraint distance, namely a front distance of the radar installation position in the detection direction, on the basis of which the front boundary of the people counting space can be determined. Preferably, a plane in the horizontal direction of the space is determined by the left boundary, the right boundary and the space front boundary, and then the space itself is based on the height constraint, thereby obtaining the demographics space.
Further, the circumferential partition angle threshold value refers to an angle division constraint step length based on one circle of rotation of the radar installation position, and the forward partition distance threshold value refers to a constraint step length based on forward regional division of the radar installation position; according to the circumferential partition angle threshold, performing circumferential cutting on the people counting space based on the radar installation position, and according to the forward partition distance threshold, performing forward cutting on the people counting space based on the boundary line where the radar installation position is located; and taking intersection of the two space division results obtained by cutting to obtain a final space region division result. It should be noted that, although the embodiment of the present application discloses a partition mode in which the circumferential partition and the forward partition are combined, other partition modes are not limited, other partition modes can be spatially partitioned, and the size of each partitioned area satisfies that the conventional partition mode only can accommodate a single function is within the protection scope of the embodiment of the present application. Illustratively, as shown in fig. 2, for a schematic view of hotel room partitioning, x1 is the left boundary constraint distance, x2 is the right boundary constraint distance, y is the front boundary constraint distance,And dividing hotel rooms by combining the threshold value of the forward dividing distance for the threshold value of the circumferential dividing angle.
S20: detecting the people counting space by using a radar to obtain a distance-Doppler spectrum; further, the radar is a millimeter wave radar.
In particular, the radar is preferably a millimeter wave radar, but is not limited to other sensors that can perform the same detection function. The distance-Doppler spectrum refers to a feature map constructed by detecting the people counting space through the radar, and the moving speed of the moving target and the distance of the moving target from the radar can be determined through the distance-Doppler spectrum, so that the moving target can be mapped to the people counting space conveniently to realize accurate people counting.
Preferably, the range-doppler profile determination procedure is a conventional procedure, as detailed below:
Transmitting radar waves to the people counting space through a millimeter wave radar, and receiving returned radar echo signals; and amplifying, mixing, low-pass filtering, digital sampling and other conventional preprocessing are carried out on the radar echo information to obtain radar baseband data. And then performing fast Fourier transform and Doppler Fourier transform on the radar baseband data in the distance dimension to obtain the distance-Doppler spectrum of all target units in the demographic space. Because the space of the embodiment of the application is two-dimensional, the target unit in the statistical space of the number of people is obtained by carrying out two-dimensional constant false alarm detection on the distance-Doppler spectrum. And the point cloud accuracy mapping in the back stepping pedestrian number statistical space is facilitated.
S30: according to the radar installation position, obtaining a target point cloud by carrying out angle estimation on a target unit detected by the distance-Doppler spectrum;
Specifically, the target point cloud refers to a set of distribution coordinates of a target unit in a demographics space determined by performing angle estimation based on the target unit detected by a range-doppler spectrum.
Preferably, the angle estimation embodiment is exemplified by the following, without limitation:
Further, according to the radar installation position, the target point cloud is obtained by performing angle estimation on the target unit detected by the distance-doppler spectrum, including:
Performing azimuth-pitch combined angle measurement on the target unit to obtain a target azimuth angle and a target pitch angle;
obtaining distance information of the target unit according to the distance-Doppler spectrum;
and determining a target distribution position according to the target azimuth angle, the target pitch angle and the distance information, and setting the target distribution position as the target point cloud.
In detail, the azimuth-elevation combined goniometer refers to a scheme of determining a distribution direction thereof based on an azimuth angle and a pitch angle of a certain target. The range-doppler spectrum can give the target azimuth: azimuth angle of target unit, target pitch angle: the pitch angle of the target unit and thus its direction. Further, the range-doppler profile may also give range information for the target unit and radar; and determining the distribution position of the target by combining the distribution azimuth and the distance information, namely, the distribution space coordinate of the target in the statistical space of the number of people. Further, analyzing all target units to obtain the target point cloud.
Further, according to the radar installation position, obtaining the target point cloud by performing angle estimation on the target unit detected by the distance-doppler spectrum includes:
Performing azimuth-pitch combined angle measurement on the target unit to obtain a target azimuth angle and a target pitch angle;
obtaining distance information of the target unit according to the distance-Doppler spectrum;
Determining a target distribution position according to the target azimuth angle, the target pitch angle and the distance information, and setting the target distribution position as an initial target point cloud;
obtaining speed information of the target unit according to the distance-Doppler spectrum;
and deleting the target distribution position of the target unit with the speed information equal to 0 from the initial target point cloud to obtain the target point cloud.
In detail, since the target point cloud is mainly used for detecting moving targets, and the accuracy of detecting static targets is not high, in order to obtain more accurate number of dynamic targets, in the embodiment of the application, the non-moving targets need to be deleted at this time, so in another embodiment, the target point cloud determined by combining azimuth-elevation combined angle measurement with distance information is regarded as an initial target point cloud. Further, velocity information of the target unit is obtained from a range-doppler spectrum, the target distribution position of the target unit with velocity information equal to 0 is deleted from the initial target point cloud, and the target point cloud is obtained.
S40: performing cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions;
Specifically, since the target point cloud may have an error due to multipath influence, it is necessary to obtain a plurality of cluster centers of the point cloud distribution by performing cluster analysis on the target point cloud, each of the cluster centers having one distribution position in the demographics space and being stored as a plurality of cluster center distribution positions.
Further, performing cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions, including:
Setting a density threshold;
And carrying out cluster analysis on the target point cloud based on a density threshold value to obtain the distribution positions of the plurality of cluster centers.
In detail, the clustering algorithm is preferably a density clustering algorithm, a density threshold is set, and point clouds are clustered based on dbscan clustering method: illustratively, adjusting each point to calculate a local density neighborhood based on dbscan clustering method, calculating the local density of each point, and when the local density is greater than or equal to a density threshold, regarding the point as a clustering center of the neighborhood; when the local density is smaller than the density threshold, the neighborhood comprises points which do not take the point as a clustering center; the clustering center is determined by finding a neighborhood as close as possible to when the local density is equal to the density threshold value through the principle. Further, the positions at which the points of the plurality of cluster centers are distributed are obtained and set as cluster center distribution positions.
S50: according to the distribution positions of the plurality of clustering centers, matching the distribution area number of the space area division result to obtain the dynamic target number;
Specifically, the plurality of cluster center distribution positions have corresponding spatial region division results, and the number of spatial region division results in which the cluster center distribution positions are statistically distributed is set as the dynamic target number. The deletion of the points with zero speed and the deletion of the interference information and the region division are realized through the above deletion of the points with zero speed and the clustering, so that the statistical accuracy is further improved, and the calculation accuracy of the number of dynamic targets is ensured.
S60: traversing the space region division result to perform inching feature detection to obtain the number of stationary targets;
Further, as shown in fig. 3, traversing the space region division result to perform inching feature detection, to obtain the number of stationary targets, including:
Detecting a plurality of groups of micro-motion features of the space region division result, wherein the micro-motion features at least comprise one or more of respiratory features and heartbeat features;
And obtaining the number of detection areas according to the groups of inching features, and setting the number of detection areas as the number of static targets.
Specifically, the micro-motion feature refers to a respiratory feature and a heartbeat feature, and can be obtained through a sound sensor, and whether any area has the respiratory feature and the heartbeat feature is detected by way of example; when there are one or more areas in the respiratory feature and the heartbeat feature and the areas which are not distributed by the dynamic targets, the areas can be divided into areas distributed by the static targets, the number of the areas which are divided into the static target distribution is counted and is set as the number of detected areas, and the number of the detected areas is set as the number of the static targets.
S70: and obtaining a statistical result of the number of people according to the number of the dynamic targets and the number of the static targets.
Further, according to the dynamic target number and the static target number, obtaining a statistical result of the number of people includes: and adding the dynamic target quantity and the static target quantity to obtain the people counting result.
Specifically, the demographics result refers to a target value obtained by detecting and counting the demographics space, and in one possible implementation, the demographics result is obtained by directly adding the number of dynamic targets and the number of static targets.
Further, according to the dynamic target number and the static target number, obtaining a statistical result of the number of people includes:
adding the dynamic target quantity and the static target quantity to set a counting result of the number of people at a first moment;
counting the number of people at a plurality of moments in the people counting space, and obtaining a second moment number counting result to an N moment number counting result, wherein the first moment and the second moment belong to a preset time zone until the N moment, N is an integer, and N is more than or equal to 1;
And carrying out concentrated trend analysis on the first time people counting result and the second time people counting result until the N time people counting result to obtain the people counting result.
Specifically, in another possible embodiment, since the number of single-time statistics may not be accurate enough, the sum of the number of dynamic targets and the number of static targets is set as the first-time demographics result, and a plurality of times in the time that subsequently satisfies the user preset time zone: continuously monitoring from the first moment to the second moment until the N moment to obtain a second moment people counting result until the N moment, and then carrying out centralized trend analysis on the first moment people counting result and the second moment people counting result until the N moment people counting result to obtain the people counting result. Preferably, the concentrated trend analysis is to delete discrete values and calculate a crowd value, and any conventional means for achieving this function can be used.
In summary, the embodiment of the application has at least the following technical effects:
1. According to the embodiment of the application, the static target quantity is determined by performing region division on the space to be detected and performing micro-motion detection; meanwhile, determining the target point cloud for weakening the influence of multipath in the people counting space by combining the distance-Doppler spectrum with angle estimation; further clustering the target point cloud to obtain the number of the region categories corresponding to the distribution positions of the plurality of clustering centers, and determining the number of dynamic targets; and obtaining a statistical result of the number of people according to the number of the dynamic targets and the number of the static targets. The detection of the static target is realized by utilizing the micro-motion feature detection, the dynamic target with higher accuracy is obtained by utilizing the combination of the distance-Doppler spectrum and the angle estimation and then clustering, and the technical effect of improving the statistical accuracy of the number of people in a specific space is achieved.
2. By fitting the people counting values at a plurality of moments, the counting fluctuation of a single time is avoided, and the stability and accuracy of the people counting result are improved.
Example two
Based on the same inventive concept as one of the spatial people counting methods in the previous embodiments, as shown in fig. 4, the present application provides a spatial people counting system, comprising:
A space acquisition module 100, configured to acquire a preconfigured demographics space, where the demographics space includes a space region division result;
the radar detection module 200 is used for detecting the people counting space through a radar to obtain a distance-Doppler spectrum;
The point cloud obtaining module 300 is configured to obtain a target point cloud by performing angle estimation on the target unit detected by the range-doppler spectrum according to the radar installation position;
The point cloud clustering module 400 is configured to perform cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions;
the dynamic target sorting module 500 is configured to obtain a dynamic target number according to the number of distribution areas of the spatial area division result matched by the distribution positions of the plurality of clustering centers;
the stationary target sorting module 600 is configured to traverse the spatial region division result to perform micro motion feature detection, so as to obtain the number of stationary targets;
The people counting module 700 is configured to obtain a people counting result according to the dynamic target number and the static target number.
Further, the people counting module 700 performs the steps of:
and adding the dynamic target quantity and the static target quantity to obtain the people counting result.
Further, the people counting module 700 performs the steps of:
adding the dynamic target quantity and the static target quantity to set a counting result of the number of people at a first moment;
counting the number of people at a plurality of moments in the people counting space, and obtaining a second moment number counting result to an N moment number counting result, wherein the first moment and the second moment belong to a preset time zone until the N moment, N is an integer, and N is more than or equal to 1;
And carrying out concentrated trend analysis on the first time people counting result and the second time people counting result until the N time people counting result to obtain the people counting result.
Further, the space obtaining module 100 performs the steps of:
Obtaining a radar installation position;
Obtaining space range basic information, wherein the space range basic information comprises a left boundary constraint distance, a right boundary constraint distance and a front boundary constraint distance;
configuring a people counting space based on the radar installation position according to the left boundary constraint distance, the right boundary constraint distance and the front boundary constraint distance;
and obtaining a circumferential partitioning angle threshold and a forward partitioning distance threshold, and partitioning the people counting space based on the radar installation position to obtain a space region partitioning result.
Further, the executing steps of the point cloud obtaining module 300 include:
Performing azimuth-pitch combined angle measurement on the target unit to obtain a target azimuth angle and a target pitch angle;
obtaining distance information of the target unit according to the distance-Doppler spectrum;
and determining a target distribution position according to the target azimuth angle, the target pitch angle and the distance information, and setting the target distribution position as the target point cloud.
Further, the executing steps of the point cloud obtaining module 300 include:
Performing azimuth-pitch combined angle measurement on the target unit to obtain a target azimuth angle and a target pitch angle;
obtaining distance information of the target unit according to the distance-Doppler spectrum;
Determining a target distribution position according to the target azimuth angle, the target pitch angle and the distance information, and setting the target distribution position as an initial target point cloud;
obtaining speed information of the target unit according to the distance-Doppler spectrum;
and deleting the target distribution position of the target unit with the speed information equal to 0 from the initial target point cloud to obtain the target point cloud.
Further, the performing steps of the point cloud clustering module 400 include:
Setting a density threshold;
And carrying out cluster analysis on the target point cloud based on a density threshold value to obtain the distribution positions of the plurality of cluster centers.
Further, the stationary object sorting module 600 performs the steps of:
Detecting a plurality of groups of micro-motion features of the space region division result, wherein the micro-motion features at least comprise one or more of respiratory features and heartbeat features;
And obtaining the number of detection areas according to the groups of inching features, and setting the number of detection areas as the number of static targets.
Further, the radar is a millimeter wave radar.
The specification and figures are merely exemplary illustrations of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (6)
1. A method of spatial people counting comprising:
Obtaining a preconfigured people counting space, wherein the people counting space comprises a space region division result;
detecting the people counting space by using a radar to obtain a distance-Doppler spectrum;
according to the radar installation position, obtaining a target point cloud by carrying out angle estimation on a target unit detected by the distance-Doppler spectrum;
performing cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions;
according to the distribution positions of the plurality of clustering centers, matching the distribution area number of the space area division result to obtain the dynamic target number;
Traversing the space region division result to perform inching feature detection to obtain the number of stationary targets;
Obtaining a statistical result of the number of people according to the number of dynamic targets and the number of static targets, wherein the statistical result comprises the following steps:
adding the dynamic target quantity and the static target quantity to set a counting result of the number of people at a first moment;
counting the number of people at a plurality of moments in the people counting space, and obtaining a second moment number counting result to an N moment number counting result, wherein the first moment and the second moment belong to a preset time zone until the N moment, N is an integer, and N is more than or equal to 1;
Carrying out centralized trend analysis on the first time people counting result and the second time people counting result until the N time people counting result to obtain the people counting result, wherein the centralized trend analysis is a process of deleting discrete values and solving a crowd value;
Wherein the obtaining a preconfigured demographic space comprises:
Obtaining a radar installation position;
Obtaining space range basic information, wherein the space range basic information comprises a left boundary constraint distance, a right boundary constraint distance and a front boundary constraint distance;
configuring a people counting space based on the radar installation position according to the left boundary constraint distance, the right boundary constraint distance and the front boundary constraint distance;
A circumferential partitioning angle threshold and a forward partitioning distance threshold are obtained, and the people counting space is partitioned based on the radar installation position, so that a space region partitioning result is obtained;
And traversing the space region division result to perform inching feature detection to obtain the number of static targets, wherein the inching feature detection comprises the following steps:
Detecting a plurality of groups of micro-motion features of the space region division result, wherein the micro-motion features at least comprise one or more of respiratory features and heartbeat features;
And obtaining the number of detection areas according to the groups of inching features, and setting the number of detection areas as the number of static targets.
2. The method of claim 1, wherein obtaining a target point cloud from radar installation locations by angle estimation of target units detected by the range-doppler profile comprises:
Performing azimuth-pitch combined angle measurement on the target unit to obtain a target azimuth angle and a target pitch angle;
obtaining distance information of the target unit according to the distance-Doppler spectrum;
and determining a target distribution position according to the target azimuth angle, the target pitch angle and the distance information, and setting the target distribution position as the target point cloud.
3. The method of claim 1, wherein obtaining a target point cloud from radar installation locations by angle estimation of target units detected by the range-doppler profile comprises:
Performing azimuth-pitch combined angle measurement on the target unit to obtain a target azimuth angle and a target pitch angle;
obtaining distance information of the target unit according to the distance-Doppler spectrum;
Determining a target distribution position according to the target azimuth angle, the target pitch angle and the distance information, and setting the target distribution position as an initial target point cloud;
obtaining speed information of the target unit according to the distance-Doppler spectrum;
and deleting the target distribution position of the target unit with the speed information equal to 0 from the initial target point cloud to obtain the target point cloud.
4. The method of claim 1, wherein performing cluster analysis on the target point cloud to obtain a plurality of cluster center distribution locations comprises:
Setting a density threshold;
And carrying out cluster analysis on the target point cloud based on a density threshold value to obtain the distribution positions of the plurality of cluster centers.
5. The method of claim 1, wherein the radar is a millimeter wave radar.
6. A space population statistics system, comprising:
the space acquisition module is used for acquiring a pre-configured people counting space, wherein the people counting space comprises a space region division result;
The radar detection module is used for detecting the people counting space through a radar to obtain a distance-Doppler spectrum;
The point cloud obtaining module is used for obtaining a target point cloud by carrying out angle estimation on the target unit detected by the distance-Doppler spectrum according to the radar installation position;
The point cloud clustering module is used for carrying out cluster analysis on the target point cloud to obtain a plurality of cluster center distribution positions;
The dynamic target sorting module is used for matching the number of the distribution areas of the space area division result according to the distribution positions of the plurality of clustering centers to obtain the number of dynamic targets;
The static target sorting module is used for traversing the space region division result to detect micro-motion characteristics and obtain the number of static targets;
the people counting module is used for obtaining a people counting result according to the dynamic target quantity and the static target quantity, and comprises the following steps:
adding the dynamic target quantity and the static target quantity to set a counting result of the number of people at a first moment;
counting the number of people at a plurality of moments in the people counting space, and obtaining a second moment number counting result to an N moment number counting result, wherein the first moment and the second moment belong to a preset time zone until the N moment, N is an integer, and N is more than or equal to 1;
Carrying out centralized trend analysis on the first time people counting result and the second time people counting result until the N time people counting result to obtain the people counting result, wherein the centralized trend analysis is a process of deleting discrete values and solving a crowd value;
Wherein the obtaining a preconfigured demographic space comprises:
Obtaining a radar installation position;
Obtaining space range basic information, wherein the space range basic information comprises a left boundary constraint distance, a right boundary constraint distance and a front boundary constraint distance;
configuring a people counting space based on the radar installation position according to the left boundary constraint distance, the right boundary constraint distance and the front boundary constraint distance;
A circumferential partitioning angle threshold and a forward partitioning distance threshold are obtained, and the people counting space is partitioned based on the radar installation position, so that a space region partitioning result is obtained;
And traversing the space region division result to perform inching feature detection to obtain the number of static targets, wherein the inching feature detection comprises the following steps:
Detecting a plurality of groups of micro-motion features of the space region division result, wherein the micro-motion features at least comprise one or more of respiratory features and heartbeat features;
And obtaining the number of detection areas according to the groups of inching features, and setting the number of detection areas as the number of static targets.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410309379.8A CN117908003B (en) | 2024-03-19 | 2024-03-19 | Space people counting method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410309379.8A CN117908003B (en) | 2024-03-19 | 2024-03-19 | Space people counting method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117908003A CN117908003A (en) | 2024-04-19 |
| CN117908003B true CN117908003B (en) | 2024-06-07 |
Family
ID=90682365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410309379.8A Active CN117908003B (en) | 2024-03-19 | 2024-03-19 | Space people counting method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117908003B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110118966A (en) * | 2019-05-28 | 2019-08-13 | 长沙莫之比智能科技有限公司 | Personnel's detection and number system based on millimetre-wave radar |
| CN111232778A (en) * | 2020-03-04 | 2020-06-05 | 日立楼宇技术（广州）有限公司 | Method and device for counting number of people in elevator car |
| CN111914793A (en) * | 2020-08-14 | 2020-11-10 | 中国工商银行股份有限公司 | Early warning method, device and equipment based on regional population |
| CN112946630A (en) * | 2021-01-27 | 2021-06-11 | 上海兰宝传感科技股份有限公司 | Personnel counting and tracking method based on millimeter wave radar |
| EP3885786A1 (en) * | 2020-03-27 | 2021-09-29 | Origin Wireless, Inc. | Method, apparatus, and system for wireless vital monitoring using high frequency signals |
| CN113835074A (en) * | 2021-08-04 | 2021-12-24 | 南京常格科技发展有限公司 | People flow dynamic monitoring method based on millimeter wave radar |
| CN113945913A (en) * | 2021-10-12 | 2022-01-18 | 南昌智能新能源汽车研究院 | Method for detecting personnel in vehicle based on millimeter wave radar point cloud |
| CN114283373A (en) * | 2021-11-16 | 2022-04-05 | 阿里巴巴达摩院(杭州)科技有限公司 | Crowd situation acquisition method, device, storage medium and system |
| CN114660592A (en) * | 2022-03-21 | 2022-06-24 | 大连理工大学 | Indoor personnel information detection method based on millimeter wave radar point cloud data |
| CN116299272A (en) * | 2023-02-22 | 2023-06-23 | 大连理工大学 | Indoor personnel counting and positioning method based on millimeter wave Lei Dadian cloud data |
| CN117196105A (en) * | 2023-09-11 | 2023-12-08 | 清华大学 | People number prediction method, device, computer equipment and storage medium |
| CN117491991A (en) * | 2023-11-02 | 2024-02-02 | 深圳市源流科技有限公司 | Indoor static personnel detection method for millimeter wave radar |
-
2024
- 2024-03-19 CN CN202410309379.8A patent/CN117908003B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110118966A (en) * | 2019-05-28 | 2019-08-13 | 长沙莫之比智能科技有限公司 | Personnel's detection and number system based on millimetre-wave radar |
| CN111232778A (en) * | 2020-03-04 | 2020-06-05 | 日立楼宇技术（广州）有限公司 | Method and device for counting number of people in elevator car |
| EP3885786A1 (en) * | 2020-03-27 | 2021-09-29 | Origin Wireless, Inc. | Method, apparatus, and system for wireless vital monitoring using high frequency signals |
| CN111914793A (en) * | 2020-08-14 | 2020-11-10 | 中国工商银行股份有限公司 | Early warning method, device and equipment based on regional population |
| CN112946630A (en) * | 2021-01-27 | 2021-06-11 | 上海兰宝传感科技股份有限公司 | Personnel counting and tracking method based on millimeter wave radar |
| CN113835074A (en) * | 2021-08-04 | 2021-12-24 | 南京常格科技发展有限公司 | People flow dynamic monitoring method based on millimeter wave radar |
| CN113945913A (en) * | 2021-10-12 | 2022-01-18 | 南昌智能新能源汽车研究院 | Method for detecting personnel in vehicle based on millimeter wave radar point cloud |
| CN114283373A (en) * | 2021-11-16 | 2022-04-05 | 阿里巴巴达摩院(杭州)科技有限公司 | Crowd situation acquisition method, device, storage medium and system |
| CN114660592A (en) * | 2022-03-21 | 2022-06-24 | 大连理工大学 | Indoor personnel information detection method based on millimeter wave radar point cloud data |
| CN116299272A (en) * | 2023-02-22 | 2023-06-23 | 大连理工大学 | Indoor personnel counting and positioning method based on millimeter wave Lei Dadian cloud data |
| CN117196105A (en) * | 2023-09-11 | 2023-12-08 | 清华大学 | People number prediction method, device, computer equipment and storage medium |
| CN117491991A (en) * | 2023-11-02 | 2024-02-02 | 深圳市源流科技有限公司 | Indoor static personnel detection method for millimeter wave radar |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117908003A (en) | 2024-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111983600B (en) | Target detection method, device and equipment | |
| CN106646338B (en) | A kind of quickly accurate indoor orientation method | |
| US20120093359A1 (en) | Batch Detection Association for Enhanced Target Descrimination in Dense Detection Environments | |
| CN111708022B (en) | Method and device for calculating scanning area boundary of millimeter wave radar | |
| CN112051568B (en) | Pitching angle measurement method of two-coordinate radar | |
| CN112689775B (en) | Radar point cloud clustering method and device | |
| US11762085B2 (en) | Device, system and method for localization of a target in a scene | |
| CN109613509A (en) | A kind of clustering method and device of trailer-mounted radar scattering point | |
| CN113075648B (en) | Clustering and filtering method for unmanned cluster target positioning information | |
| CN113687429B (en) | Device and method for determining boundary of millimeter wave radar monitoring area | |
| WO2019200153A1 (en) | System and method of angle-of-arrival estimation, object localization, and target tracking, with received signal magnitude | |
| CN115546526B (en) | Three-dimensional point cloud clustering method, device and storage medium | |
| EP1269214A2 (en) | Localisation of a signal emitting source | |
| CN110837079A (en) | Target detection method and device based on radar | |
| CN114690174A (en) | Target tracking method and device based on millimeter wave radar and laser radar | |
| CN117908003B (en) | Space people counting method and system | |
| CN114578341A (en) | Radar tracking method, noise elimination method, device and equipment | |
| US10012729B2 (en) | Tracking subjects using ranging sensors | |
| CN114219908A (en) | Map construction method and device | |
| CN113820704A (en) | Method and device for detecting moving target and electronic equipment | |
| CN111722196A (en) | Radar reflection point extraction method and device | |
| Mikhalev et al. | Fusion of sensor data for source localization using the Hough transform | |
| US20240255633A1 (en) | Apparatus and method for tracking objects | |
| CN117872354B (en) | Fusion method, device, equipment and medium of multi-millimeter wave Lei Dadian cloud | |
| CN118376994B (en) | Target positioning method based on multi-radiation-source range Doppler image strong point estimation |
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 | ||
| GR01 | Patent grant |