Human gait analysis using wavelet de-noising and total variation filtering

This paper analyses the walking gait of a person in time domain, frequency domain and time-frequency domain for counting of steps. The human gait movement is the primary source of movement for humans. This movement is one of the primary causes of energy consumption for the human body. While it is one of the most common features of living beings, it is complex to understand and analyze. The human gait movement consists of cycles of repeated motion. These repeated patterns can be analyzed and understood in order to extract different features of walking. In this paper, human gait movement data is acquired with a foot mounted accelerometer followed by detailed decomposition and extraction of patterns by Fourier Transform, FFT and WT.