Behavior intention open set identification method and system based on neural network, and electronic equipment

The invention discloses a behavior intention open set recognition method and system based on a neural network and electronic equipment, and belongs to the field of intention recognition. According to the method, firstly, a known class classification network comprising a self-attention hybrid convolu...

Full description

Saved in:
Bibliographic Details
Main Authors FANG SONGSONG, SHENG QUAN, SUN XIAOXIAO, YE CHUNYI
Format Patent
LanguageChinese
English
Published 05.07.2024
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The invention discloses a behavior intention open set recognition method and system based on a neural network and electronic equipment, and belongs to the field of intention recognition. According to the method, firstly, a known class classification network comprising a self-attention hybrid convolution module and a time sequence change sensing module is adopted, key local features in the time dimension and global time sequence change are extracted respectively, and the key local features and the global time sequence change are processed into low-dimensional class activation vectors; and fitting Weibull distribution of each action intention category and an unknown action intention category judgment threshold value in a training process through an adaptive threshold value detection module, correcting a category activation vector of a test sample in a test or reasoning stage, and identifying the sample of which the threshold value is smaller than the threshold value of the corresponding category as the unknown
Bibliography:Application Number: CN202410471191