Human body daily behavior activity recognition optimization method based on stacking noise reduction auto-encoder

Abstract

The invention relates to a human body daily behavior activity recognition optimization method based on a stacked noise reduction auto-encoder, which comprises the following steps: 1) based on a deep learning method and an SDAE model, constructing a classification model of human body behavior recognition by adopting human body behavior data acquired from a wearable sensor; 2) initializing a classification model for the data collected by the wearable sensor, and dividing the initialized data into a training set, a test set and a verification set; 3) carrying out pre-training and fine tuning processing on the constructed classification model; and 4) verifying the generalization ability of the trained classification model by using the verification set in a cross verification mode, and inputting the test set into the classification model to obtain the corresponding prediction category. Compared with the prior art, the method has the advantages that a more accurate model identification result is obtained, more directions can be provided for the application of behavior identification, and the like.

Description

technical field [0001] The present invention relates to the technical field of human behavior activity recognition based on wearable devices, in particular to a human daily behavior activity recognition optimization method based on stacked noise reduction autoencoders. Background technique [0002] Behavior recognition technology has received more and more attention because of its wide range of applications. Behavior recognition technology has more and more applications in industry, health monitoring, and indoor positioning. Human behavior can be classified from different angles. According to the characteristics of the behavior, it can be divided into static behavior, dynamic behavior and transfer behavior. Among them, static behavior and dynamic behavior are the focus of current researchers, while transfer behavior is often ignored. Transfer behavior It also plays an important role in the medical field. The transfer behavior is an intermediate activity between two static or...

AI Technical Summary

Problems solved by technology

Machine learning methods have achieved good results in the field of behavior recognition, but there is still a defect in machine learning methods: when performing feature extraction, researchers often need to manually extract features, which leads to a lot of time and energy consumption

The existing technology uses the SDAE model to identify human movement behavior, and has achieved good accuracy, but this method also does not consider the application of transfer behavior in the field of behavior recognition

Images