Abstract:
Age related macular degeneration (ARMD) is a medical condition which results in
deterioration of human retina and in particularly macula. It is caused due to deposits of
drusen on the retina and the disease may cause severe blindness. It is important to detect
ARMD in its early stages to save patient’s vision. In this project, a new technique for drusen detection from fundus images by using Gabor kernel based filter bank is used. Spurious regions are eliminated which may be confused with drusen. The system consists of phases like preprocessing, bright region detection, optic disc detection, removal of optic disc and finally AMD detection using drusen regions. The system is evaluated by testing it on STARE database using performance factor like accuracy. We have achieved 98% accuracy on this system. The results show the comparison and validity of our system with existing techniques.
