Main Article Content
Electroencephalogram (EEG) signals help us to analyse the various activities of a human brain. These signals reveals the excellent activity of a brain at certain states and these neuroimaging methods differs from other neuroimaging methods such as magnetoencephalogram, functional magnetic resonance imaging, Positron emission tomography by its capabilities like high temporal resolution in the millisecond range, low cost, portability and non-invasiveness. The patterns that are recorded by these EEG signals are mostly non-stationary, time and frequency variant type and with the increasing power of computing and enhanced processing capabilities of the recent tools, EEG signal analysis can be done efficiently and effectively. In recent days, EEG signal analysis through the classification models is utilized in different application areas like diagnosis of various neurological disorders in the medical field, emotion recognition, motor imagery and entertainment. A variety of signal processing techniques must be used to process such signals.This review offers an extensive study that explores the various stages of processing EEG signals such as data acquisition, pre-processing techniques which include artifacts removal, feature extraction methods of different domains, post processing techniques and the classification models in accordance with the different applications that utilize EEG signals.