Ship Silhouette Recognition Using Principal Components Analysis

We report on an evaluation study of a ship classi er based on the Principal Components Analysis (PCA). A set of ship pro les are used to build a covariance matrix which is diagonalized using the Karhunen-Lo eve transform. A subset of the principal components corresponding to the highest eigenvalues are selected as the ship features space. The recognition process consists in projecting a pro le on this eigen-subspace and performing a similarity measure (herein a standard Euclidean distance). We have measured the recognition performance of the classi er using various sets of range-pro le signatures of ship silhouette images and simulated synthetic aperture radar images of ships under various aspect angles. It is found that the PCA-based ship classi er design o ers good class discriminacy when trained with a limited number of ship classes (< 10) under an aspect angle range of 60 degrees about the ship side view. Additional tests are however necessary to validate the classi er on larger data sets and real images.