Leveraging Ensemble Learning Model for Human Gait Recognition in Human Computer Interface

Human Gait Recognition (HGR) is a biometric system that is exploited for security purpose over the past few years. The efficiency of gait detection is affected by different features namely walking surfaces, wearing clothes, and carrying a bag. Moreover, detection from divergent views is a complicated problem in HGR. Numerous approaches were proposed for HGR using classical and DL methods but the conventional technique is not appropriate for massive data. This study leverages Ensemble Learning Model for Human Gait Recognition in Human Computer Interface (ELMHGA-HCI) technique. The ELMHGAHCI technique intends to recognize various kinds of human gaits to assist HCI applications. In the ELMHGA-HCI technique, Time Domain (TD) and Frequency Domain (FD) features from the input signals. For gait recognition, an ensemble of three DL models takes place namely Bidirectional Long Short Term Memory (BiLSTM), Gated Recurrent Unit (GRU) and Recurrent Neural Network (RNN). To exhibit the enhanced gait recognition results of the ELMHGA-HCI technique, a widespread simulation analysis is done. The simulation outcomes pointed out the enhanced performance of the ELMHGA-HCI approach over other existing techniques.