Abstract: Wire rope is more and more widely used in industrial production. Its safety has become increasingly prominent. In the field of wire rope nondestructive testing, magnetic flux leakage testing is the most mature and widely used method. However, it is difficult to extract defect features directly from the signals of magnetic flux leakage testing due to the influence of testing environment. Therefore, a defect detection method based on empirical mode decomposition is proposed to detect defects. Firstly, the noise component in magnetic flux leakage signal is removed via wavelet and adaptive soft threshold method. Then, the obtained signal is decomposed by empirical mode decomposition, and the weakly related components are further processed by the wavelet method to extract the weak useful information. Finally, the signal is reconstructed by combining the weak information with the moderately related and above components, thus extracting the damage characteristics and determining damage types and damage locations. The proposed method achieves the maximum utilization of the wire rope in the safety available range with a good economic benefits.