CN-121971055-A - Identity determination method and device, storage medium and electronic device

CN121971055ACN 121971055 ACN121971055 ACN 121971055ACN-121971055-A

Abstract

The application discloses an identity determining method and device, a storage medium and an electronic device, and relates to the technical field of smart families, wherein the identity determining method comprises the steps of acquiring N echo signals acquired by a radar positioned at a target position, and determining a three-dimensional signal matrix according to the N echo signals, wherein N is an integer greater than or equal to 2, and the dimensions of the three-dimensional signal matrix comprise a distance dimension, an angle dimension and a Doppler frequency dimension; the method comprises the steps of determining a target feature vector of a target object located at a target position according to a three-dimensional signal matrix, wherein vector elements in the target feature vector comprise radar scattering sectional areas, barycenter distances, barycenter angles, respiratory frequencies, heart rate frequencies and phase differences among N radars of the target object, and determining target identity marks of the target object according to the target feature vector of the target object based on a preset feature vector library. By adopting the technical scheme, the problem that the identity of the target object cannot be accurately identified is solved.

Inventors

CUI ZHEN
HAO SHUHUA
MA CHENGDONG
TIAN YUNLONG
ZHANG HUAJUN
NIU LI
ZHAO XIANGYANG
WANG SHENGJIE

Assignees

青岛海尔科技有限公司
海尔优家智能科技（北京）有限公司
青岛海尔智能家电科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260402

Claims (10)

1. An identity determination method, comprising: Acquiring N radar acquired echo signals at a target position, and determining a three-dimensional signal matrix according to the N echo signals, wherein N is an integer greater than or equal to 2, and the dimensions of the three-dimensional signal matrix comprise a distance dimension, an angle dimension and a Doppler frequency dimension; Determining a target feature vector of a target object positioned at the target position according to the three-dimensional signal matrix, wherein vector elements in the target feature vector comprise radar scattering sectional area, barycenter distance, barycenter angle, respiratory frequency and heart rate frequency of the target object and phase differences among the N radars; And determining a target identity of the target object according to the target feature vector of the target object based on a preset feature vector library, wherein the preset feature vector library is provided with a plurality of feature vectors stored in advance and identities corresponding to the feature vectors.
2. The method according to claim 1, wherein determining the target identity of the target object based on the target feature vector of the target object based on the preset feature vector library, comprises: Determining cosine similarity between the target feature vector and each feature vector in the preset feature vector library; And determining the identity of the feature vector with the cosine similarity with the target feature vector in the preset feature vector library being greater than a preset threshold value as the target identity.
3. The identity determination method according to claim 2, wherein after determining the target identity of the target object according to the target feature vector of the target object based on a preset feature vector library, the method further comprises: updating a sample feature vector through the following formula, wherein the sample feature vector is a feature vector with cosine similarity with the target feature vector in the preset feature vector library being larger than a preset threshold value: ; Wherein, the For the feature vector of the sample, And k is the number of times of cosine similarity calculation with the sample feature vector for the target feature vector.
4. An identity determination method according to claim 1, wherein before determining a three-dimensional signal matrix from N echo signals, the method further comprises: Collecting the vibration frequency of equipment in the space where the target object is located; Generating a reverse vibration signal according to the vibration frequency, and adding the reverse vibration signal to each of the N echo signals.
5. An identity determination method according to claim 1, wherein before determining a three-dimensional signal matrix from N echo signals, the method further comprises: and performing adaptive notch filtering processing on each of the N echo signals, wherein the notch center frequency comprises 50Hz, 900MHz and 1800MHz, and the radar is positioned in an electromagnetic shielding cover.
6. The method according to claim 1, wherein after determining the target identity of the target object according to the target feature vector of the target object based on a preset feature vector library, the method further comprises: determining the body movement times of the target object according to the signal amplitude change of the signal corresponding to the target identity in the N echo signals; determining the turning frequency of the target object according to the angle change corresponding to the target identity in the three-dimensional signal matrix; Under the condition that a reference feature vector is not determined according to a three-dimensional signal matrix determined by echo signals continuously collected by the N radars, determining that the target object leaves the target position, wherein the cosine similarity between the reference feature vector and the target feature vector is larger than a preset threshold; And determining the respiratory frequency and the heart rate of the target object according to the Doppler frequency corresponding to the target identity in the three-dimensional signal matrix.
7. An identity determining apparatus, comprising: the first determining module is used for acquiring N echo signals acquired by the radar at the target position and determining a three-dimensional signal matrix according to the N echo signals, wherein N is an integer greater than or equal to 2, and the dimensions of the three-dimensional signal matrix comprise a distance dimension, an angle dimension and a Doppler frequency dimension; The second determining module is used for determining a target feature vector of a target object positioned at the target position according to the three-dimensional signal matrix, wherein vector elements in the target feature vector comprise radar scattering cross section area, barycenter distance, barycenter angle, respiratory frequency and heart rate frequency of the target object and phase differences among the N radars; The third determining module is configured to determine, based on a preset feature vector library, a target identity of the target object according to a target feature vector of the target object, where the preset feature vector library has a plurality of feature vectors stored in advance and identities corresponding to each feature vector.
8. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 6.
9. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to execute the method according to any of claims 1 to 6 by means of the computer program.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method as claimed in any one of claims 1 to 6.

Description

Identity determination method and device, storage medium and electronic device Technical Field The application relates to the technical field of smart families, in particular to an identity determining method and device, a storage medium and an electronic device. Background In the fields of home care and intelligent health monitoring, a non-contact sleep state sensing technology gradually replaces traditional wearable equipment, and the non-contact sleep state sensing technology becomes a main stream direction of non-inductive physiological and behavior monitoring. Millimeter wave radars are widely applied to bedside human activity monitoring due to the advantages of strong penetrability, good privacy protection, no influence of illumination and the like. In the prior art, radar echo signals are generally collected, and human body micro-motion information is obtained through three-dimensional transformation of distance, angle and Doppler, so that physiological parameters such as respiration, heart rate and body movement are analyzed. However, the current mainstream scheme generally lacks stable identification capability for individual identity in a multi-person co-sleeping scenario. Although different targets can be roughly distinguished through the characteristics of signal amplitude, position change and the like, because a personalized identity characterization mechanism based on multidimensional space motion characteristics is not established, when multiple human bodies move frequently, gesture overlap or are adjacent in position, the system is extremely easy to confuse the characteristic vectors of different individuals, so that identity misjudgment or continuity loss is caused. Therefore, how to realize continuous identification of individual identities with high robustness and high accuracy on the premise of not depending on cameras, wearable devices or active interaction of users becomes a key bottleneck for restricting deep application of millimeter wave radars in multi-user home scenes. Aiming at the problem that the identity of a target object cannot be accurately identified in the related art, no effective solution is proposed at present. Accordingly, there is a need for improvements in the related art to overcome the drawbacks of the related art. Disclosure of Invention The embodiment of the invention provides an identity determining method and device, a storage medium and an electronic device, which are used for at least solving the problem that the identity of a target object cannot be accurately identified. According to one aspect of the embodiment of the invention, an identity determination method is provided, which comprises the steps of obtaining N radar-acquired echo signals at a target position, determining a three-dimensional signal matrix according to the N echo signals, wherein the dimensions of the three-dimensional signal matrix comprise a distance dimension, an angle dimension and a Doppler frequency dimension, determining a target feature vector of a target object at the target position according to the three-dimensional signal matrix, wherein vector elements in the target feature vector comprise radar scattering cross section areas, barycentric distances, barycentric angles, respiratory frequencies, heart rate frequencies of the target object and phase differences among the N radars, and determining target identity identifiers of the target object according to the target feature vector of the target object based on a preset feature vector library, wherein the preset feature vector library is provided with a plurality of feature vectors stored in advance and identity identifiers corresponding to each feature vector. In an exemplary embodiment, determining the target identity of the target object according to the target feature vector of the target object based on a preset feature vector library includes determining cosine similarity between the target feature vector and each feature vector in the preset feature vector library, and determining the identity of the feature vector in the preset feature vector library, the cosine similarity of which is greater than a preset threshold, as the target identity. In an exemplary embodiment, after determining the target identity of the target object according to the target feature vector of the target object based on a preset feature vector library, the method further comprises updating a sample feature vector by the following formula, wherein the sample feature vector is a feature vector in the preset feature vector library, and the cosine similarity between the sample feature vector and the target feature vector is greater than a preset threshold value: Wherein, the method comprises the steps of, For the feature vector of the sample,And k is the number of times of cosine similarity calculation with the sample feature vector for the target feature vector. In an exemplary embodiment, before determining the three-dimensional signal matrix fro