MFL Data Feature Extraction based on KPCA-BOMW Model

Magnetic Flux Leakage (MFL) inspection for submarine pipelines is the most common non-destructive testing method currently. MFL data feature extraction is a vital part in MFL data processing. In order to deal with this problem, this paper proposes a MFL data feature extraction algorithm based on improved KPCA-BOMW model. This algorithm processes the data matrixes directly, chooses SURF to instead the SIFT feature in traditional BOW model to appropriate the MFL data, and uses KPCA to reduce the dimension of SURF to increase the running speed. Finally, verify the effectiveness of the algorithm by three experiments.