Optimization of retinal blood vessel segmentation based on Gabor filters and particle swarm optimization

The structure of the retinal blood vessels can be obtained by segmenting fundus images. A image gained through color photography or fluorescein angiography (FA). produced camera cause noise which reduce quality image. To noise, this research uses non-local means filter (NLMF). For texture analysis, study Gabor filters due to frequencies as same human visual system. process vessel is performed using K-means optimized particle swarm optimization (PSO). accuracy 0.9525, precision 0.8330, sensitivity 0.5817, and specificity 0.9880 are proposed method.