Application of Gabor transform as texture discriminator of masses in digital mammograms

It has been demonstrated that the class of the Gabor functions possesses the property of simultaneously minimizing uncertainty in spatial position and spatial frequency. A technique of detecting masses in digital mammogram has been investigated using Gabor Transform as a texture discriminator. The digital mammograms is "transformed" via Gabor function into a frequency spectrum representation that provides a means of discrimination between features base on the frequency differences. In addition, Gabor function are used as a model of the receptive fields of neurons in the visual cortex. The application of this technique is demonstrated using computer generated images and digital mammograms.