Non-contact personnel identity and posture recognition method based on motion instability compensation

By using motion instability compensation and GesAuthNet dual-branch deep neural network, the problem of signal interference of millimeter-wave radar on mobile platforms is solved, achieving high-precision identity authentication and gesture recognition with strong stability and robustness.

CN122110050AActive Publication Date: 2026-05-29NANJING UNIV OF POSTS & TELECOMM
14 Cites 0 Cited by

Patent Information

Application Number
CN202610574933.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-28
Publication Date
2026-05-29
Estimated Expiration
2046-04-28

AI Technical Summary

Technical Problem

Existing millimeter-wave radar-based solutions struggle to achieve joint inference of human behavior and identity on mobile platforms, and are also subject to signal interference and privacy breaches.

Method used

By using motion instability compensation technology, background point clouds are screened and radar self-motion is estimated. A GesAuthNet dual-branch deep neural network is built to extract identity and pose features, eliminate signal interference, and achieve identity authentication and gesture recognition.

Benefits of technology

It achieves high-precision identity authentication and gesture recognition on mobile platforms, with motion instability compensation capabilities and robustness, and can accurately identify personnel identity and gesture movements.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention belongs to the field of millimeter wave radar sensing and human body posture recognition, and discloses a non-contact personnel identity and posture recognition method based on motion instability compensation, which comprises the following steps: step 1, estimating a radar self-motion speed through self-adaptive screening of static background points, and compensating point cloud coordinates and a micro-Doppler spectrum; step 2, establishing a GesAuthNet double-branch deep neural network to obtain morphological parameters and posture parameters; and 3, constructing an identity authenticator in combination with the morphological parameters to authenticate the legality of the user, and constructing a gesture classifier in combination with the posture parameters to recognize the gesture of the user. The method has the advantages of motion instability compensation capability, robust gesture feature extraction capability, high-precision identity authentication and gesture recognition, and strong stability and robustness.
Need to check novelty before this filing date? Find Prior Art