Skip to Main Content
A semi-automated method for detecting eyelid-related pathologies is presented. The authors describe a novel approach for extracting significant features of the palpebral contour of the human eye in order to assist eyelid-related pathologic diagnosis. Image-processing techniques are used to better perform palpebral morphology analysis and quantification. On its practical side, it consists of a semi-automated method to detect eyelid contour using a Laplacian-of-Gaussian operator, analyze it, and treat its parameters. The main advantage of this approach is to avoid subjectivity of the traditional clinical approach.