The present work focuses on the development of a novel computer-based approach for tear ferning (TF) featuring. The original TF images of the recently developed five-point grading scale have been used to assign a grade for any TF image automatically.