On Image Analysis by the Methods of Moments
IEEE Transactions on Pattern Analysis and Machine Intelligence
C4.5: programs for machine learning
C4.5: programs for machine learning
Digital image processing
| Hi-index | 0.00 |
This paper presents a series of experiments testing thefeasibility of employing image-processing techniquesfor the feature extraction stage in the implementation ofa basic optic nerve image classifier. Such a schemecompletely removes the need for manually identifyingthe edge of the optic nerve. In this work, Zernike momentsare extracted from Confocal Scanning LaserTomography images of optic discs for the purposes ofclassifying the disc as healthy or damaged using a lineardiscriminant function derived from a linear perceptron.Our preliminary results, when compared with the performanceof conventional feature sets, demonstrate theappropriateness of this approach.