Layer extraction in rodent retinal images acquired by optical coherence tomography

Optical coherence tomography (OCT) is a modern technique that allows for in vivo, fast, high-resolution 3D imaging. OCT can be efficiently used in eye research and diagnostics, when retinal images are processed to extract borders of retinal layers. In this paper, we present two novel algorithms for delineation of three main borders in rodent retinal images. The first, fast algorithm is based on row projections in a sliding window. It provides initial borders for a slower but more precise variational algorithm that iteratively refines the borders. The results obtained by the two algorithms are quantitatively evaluated by comparison to the borders manually extracted in a set of retinal images.