Skip to Main Content
This study reports an experimental analysis of footsteps as a biometric. The focus here is on information extracted from the spatial domain of signals collected from an array of piezoelectric sensors. Results are related to the largest footstep database collected to date, with almost 20 000 valid footstep signals and more than 120 persons. A novel feature approach is proposed, obtaining three-dimensional images of the distribution of the footstep pressure along the spatial course. Experimental work is based on a verification mode with a holistic approach based on principal component analysis and support vector machines, achieving results in the range of 6-10% equal error rate (EER) depending on the experimental conditions of quantity of data used in the client models (200 and 40 signals per model, respectively). Also, this study includes the analysis of two interesting factors affecting footstep signals and especially spatial domain features, namely, sensor density and the special case of high heels.