In this paper, we propose a hybrid heading estimation scheme exploiting smart-phone inertial sensors, accelerometer, magnetic, gyroscope, for the pedestrian dead reckoning (PDR). Generally the heading estimation of a pedestrian determines the performance of indoor navigation system (INS). Even though various heading estimation schemes are known, most systems are simply based on only one scheme. Based on the performance evaluation of each scheme, in this paper, a hybrid heading estimation scheme is proposed to improve the performance of INS. To verify the effectiveness of our proposed scheme, we performed experiments in indoor environments and compared with various heading estimation schemes with a step detection scheme based on a peak detection scheme. According to the experiment results, the proposed hybrid heading estimation can provide enhanced performance than the conventional schemes.