Data identification method, device and equipment and computer readable storage medium

Abstract

The invention discloses a data identification method, device and equipment and a computer readable storage medium. The method comprises the steps of: obtaining to-be-identified spatio-temporal data, and converting the to-be-identified spatio-temporal data into a spatial covariance matrix; calling a manifold dimension reduction layer in a preset target identification model to perform dimension reduction processing on the spatial covariance matrix to obtain a dimension-reduced covariance matrix of a preset dimension; calling a tangent space projection layer in the target identification model toproject the dimension-reduced covariance matrix to a tangent space of the dimension-reduced covariance matrix to obtain a projection result; and inputting the projection result into a neural network layer in the target identification model to obtain a identification result of the to-be-recognized spatio-temporal data. A scheme for realizing a data identification task by adopting a deep learning technology is provided, so that the identification accuracy of the spatio-temporal data is improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a data identification method, device, equipment and computer-readable storage medium. Background technique [0002] In various application fields, data has become the key to solving problems. With the technological development of data acquisition equipment such as sensors, the value of spatiotemporal data collected by data acquisition equipment in specific problems is getting higher and higher. Spatiotemporal data refers to data with spatiotemporal attributes, for example, radar data collected by lidar, electroencephalogram (EEG) data collected in the field of brain-computer interface, etc. At present, deep learning has been widely used in various fields to solve various problems. However, when applied to spatiotemporal data recognition tasks, for example, it is applied to the recognition of EEG data to control external devices according to the recognized ...

AI Technical Summary

Problems solved by technology

At present, deep learning has been widely used in various fields to solve various problems. However, when applied to spatiotemporal data recognition tasks, for example, it is applied to the recognition of EEG data to control external devices according to the recognized action intentions of people. Due to the difficulty in collecting spatio-temporal data, it is difficult to meet the needs of deep learning that requires large-scale training data sets to complete model training, which has led to the lack of solutions for using deep learning technology to achieve data recognition tasks

Images