In this paper, we propose a method for human detection in the non-stationary clutter environment using the ultra-wideband (UWB) impulse radar. The foliage swaying in the wind is specifically considered as the non-stationary clutter. The breathing human and the non-stationary foliage have different characteristics of movement which results in different types of the energy variations. The method based on the excess kurtosis of the target energy distribution is used to distinguish the breathing human from the non-stationary foliage clutter. The experimental results show the performance of the proposed method.