A method was presented based on 1.5-dimensional energy spectrum. This method combines 1.5-dimensional spectrum and teager energy operator. To solve the fault diagnosis problem of rolling bearings and improve the signal to noise ratio, a new diagnosis method based on EEMD de-noising and 1.5-dimensional energy spectrum was proposed. Firstly, the fault signal was decomposed by ensemble empirical mode decomposition method and a set of intrinsic mode function components were obtained. Secondly, filtered the components by correlation coefficient - kurtosis criteria and the selected IMF components were used to reconstruct signal. Finally, in order to get bearing fault characteristic frequency information, the 1.5-dimensional energy spectrum of reconstructed signal was calculated. Simulation inner fault signal and measured inner fault signal of rolling bearings were analyzed by the proposed method, and the diagnosis results were satisfactory.