Seismocardiography (SCG) measures the precordial vibrations using a sensor called accelerometer, which is of small size and low weight. These features support better attachment of it to the subject's body and hence get less affected by the slow motion of the subject. However, noise generated due to the footsteps, while walking, contaminates the SCG signal. Therefore, in this paper, a novel method is proposed to remove these contaminations from the SCG signal. A three-Axis accelerometer was attached to the chest wall such that heart sound components are reflected in the z-Axis while the motion noise due to footstep will be seen in the x-Axis. To remove the noise components from the heart signal (z-Axis), the location of footsteps from the x-Axis are identified and the corresponding components from the z-Axis are removed. After noise removal fundamental heart sounds (FHS), S1 and S2, are identified using Otsu's threshold method. The obtained results show that the proposed algorithm outperforms the state-of-The-Art methods. It efficiently removes most of the contamination due to footsteps and identifies the heart sound components. © 2016 IEEE.