Modified nearest neighbor fuzzy classification algorithm for ship target recognition

Abstract

Modified nearest neighbor fuzzy classification (MNNFC) algorithm is proposed for the character of ship target high resolution range profile (HRRP). Ship length, dispersant, symmetry and central moments features are some stable features for ship HRRP and extracted accurately. Modified nearest neighbor fuzzy classification algorithm is designed for different features to contribute their predominance because the significance and stability of each feature are different. And the membership degree of each feature is modified differently. Experimental results with the actual measured data of 10 ships show that the proposed algorithm is very useful in ship target classification.