A Hybrid Posture Detection Framework: Integrating Machine Learning and Deep Neural Networks

The posture detection received lots of attention in the fields human sensing and artificial intelligence. Posture can be used for monitoring health status elderly remotely by identifying their postures such as standing, sitting walking. Most current studies traditional machine learning classifiers to identify posture. However, these methods do not perform well detect accurately. Therefore, this study, we proposed a novel hybrid approach based on (i. e., support vector (SVM), logistic regression (KNN), decision tree, Naive Bayes, random forest, Linear discrete analysis Quadratic analysis) deep 1D-convolutional neural network (1D-CNN), 2D-convolutional (2D-CNN), LSTM bidirectional LSTM) detection. uses prediction (ML) (DL) improve performance ML DL algorithms. experimental results widely benchmark dataset are shown achieved an accuracy more than 98%.