Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
RADAR SIGNAL-BASED IDENTITY AND GESTURE RECOGNITION METHOD
Document Type and Number:
WIPO Patent Application WO/2021/068470
Kind Code:
A1
Abstract:
A radar signal-based identity and gesture recognition method, comprising: reading radar echo signals (S1) received by a radar sensor and obtained via reflection by various gestures of a tester; mixing (D) the radar echo signals with radar emission signals; after filtering (M) and centralizing (M') the mixed signals, respectively performing training to obtain a neural network (A) capable of performing identity verification and a neural network (B) capable of performing gesture recognition. During a real-time detection process, a user identity is first verified, and if the identity verification is passed, a user gesture is valid, and the user gesture is verified and a corresponding operation is performed according to corresponding relationships between user gestures and operations. The method can simultaneously perform gesture recognition and identity verification by means of a single gesture operation, and according to whether a gesture belongs to a corresponding user, can determine whether to perform an operation corresponding to the gesture, thereby increasing the convenience and security of gesture recognition.

More Like This:
WO/2008/077980METHOD AND SYSTEM FOR DETERMINING POSITION AND SPEED OF AN ELEVATING APPARATUS
WO/1994/029744SYSTEM FOR EVALUATING THE INNER MEDIUM CHARACTERISTICS OF NON-METALLIC MATERIALS
Inventors:
WANG YONG (CN)
CHEN JUNYI (CN)
CAO JIAHE (CN)
Application Number:
PCT/CN2020/083989
Publication Date:
April 15, 2021
Filing Date:
April 09, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV ZHEJIANG (CN)
International Classes:
G01S13/02; G01S7/35; G01S7/41; G01S13/42; G01S13/50; G01S13/58
Foreign References:
CN107024685A2017-08-08
CN108469611A2018-08-31
CN110456320A2019-11-15
CN109271838A2019-01-25
CN106446828A2017-02-22
CN109188414A2019-01-11
CN108344995A2018-07-31
US20160259037A12016-09-08
CN110765974A2020-02-07
CN106537173A2017-03-22
Other References:
WANG, YONG, WANG SHA-SHA, TIAN ZENG-SHAN, ZHOU MU, WU JIN-JUN: "Two-Stream Fusion Neural Network Approach for Hand Gesture Recognition Based on FMCW Radar", ACTA ELECTRONICA SINICA, vol. 47, no. 7, 31 July 2019 (2019-07-31), pages 1408 - 1415, XP055798922
Attorney, Agent or Firm:
HANGZHOU QIUSHI PATENT OFFICE CO., LTD. (CN)
Download PDF:
View/Download PDF      PDF Help



 
Previous Patent: QUANTIZATION AND FIXED-POINT FUSION METHOD AND APPARATUS FOR NEURAL NETWORK

Next Patent: METHOD FOR DETERMINING STALLED STATE OF PERMANENT MAGNET SYNCHRONOUS MOTOR UNDER POSITION SENSOR-FRE...