Segmentation of vessel-like patterns using mathematical morphology and curvature evaluation
IEEE Transactions on Image Processing
Hi-index | 0.00 |
Human retina is a well-known biometric trait and its use is emerging as retinal blood vessels have high degree of uniqueness. To implement a retinal biometric system we need to separate the blood vessel structure. A retina segmentation system based on vector quantization using Kekre's Fast Codebook Generation Algorithm (KFCG) is proposed in this paper. The retina image is used to generate the initial codevector in a multidimensional vector space. KFCG is then applied to form clusters of the codevectors. Each cluster corresponds to an image plane. These planes are separated for blood vessel segmentation. The results are promising and this technique can be used for retina image preprocessing.