Sparse Representation and Dictionary Learning as Feature Extraction in Vessel Imagery.

摘要

This report describes experiments designed to evaluate the usefulness of a specific algorithm for classifying images of commercial ships by class. This algorithm uses a technique known as sparse coding to represent images for classification. The sparse coding algorithm is compared with another algorithm evaluated in previous publications. The sparse coding algorithm is shown to perform approximately as well as the algorithm it is compared with and does not appear to offer any improvement. Additional research is required to identify algorithms best suited for the ship classification task.