Solving basic and advanced human activities using LSTM

Because of the increasing use of numerous sensors, human activity recognition has grown popular in a variety of fields. It has been noticed that when the requirement for protection, defense, and human activity classification and recognition grows, more research in these areas becomes necessary. Such technologies are capable of detecting terrorists and assisting in disaster relief. The goal of this research study is to look at human activity recognition systems that use deep learning. In addition, the integration of such deep learning mechanisms has been assessed. It has been seen that in previous studies, data augmentation mechanisms have been utilized to perform human activity recognition, which has resulted in a significant increase in accuracy. This accuracy can be obtained with the support of multilayer LSTM network. An approach for assessing individual activity based on electro-magnetic radiation and deep learning algorithms has been reported in a number of papers. Time and space complexity are two major hurdles in such study. As a result, the goal of this analysis review is to research the technological and economic practicability of applying deep-learning for human action recognition systems. In this study, LSTM is applied for basic and advanced human activities identification.