To solve the problem that bearing early fault features are difficult to be extracted under strong background noise,an early fault diagnosis method based on the sparrow search algorithm (SSA) and the modified Laplacian of Gaussian (MOLG) filter combined with spectral amplitude modulation (SAM) was proposed.The maximum kurtosis of the filtered signal was taken as the index,and the optimal parameters of filter order and standard deviation were found adaptively by using SSA.The fault signal was filtered by using MLOG filter to remove noise.The SAM method was used to assign different weights to the amplitude of the filtered signal for reconstruction,the square envelope spectrum of the reconstructed signal was calculated,and the fault characteristic frequency was extracted.The simulation signals and real data from different test stations were verified.The comparison between PSO and GA optimized MLOG results show that the proposed SSA-MLOG-SAM method is effective and feasible for early fault feature extraction of rolling bearings.