Peer Reviewed

1

DOI

10.17077/omia.1047

Conference Location

Athens, Greece

Publication Date

October 2016

Abstract

In this paper, we propose two vesselness maps and a simple to difficult learning framework for retinal vessel segmentation which is ground truth free. The first vesselness map is the multiscale centrelineboundary contrast map which is inspired by the appearance of vessels. The other is the difference of diffusion map which measures the difference of the diffused image and the original one. Meanwhile, two existing vesselness maps are generated. Totally, 4 vesselness maps are generated. In each vesselness map, pixels with large vesselness values are regarded as positive samples. Pixels around the positive samples with small vesselness values are regarded as negative samples. Then we learn a strong classifier for the retinal image based on other 3 vesselness maps to determine the pixels with mediocre values in single vesselness map. Finally, pixels with two classifier supports are labelled as vessel pixels. The experimental results on DRIVE and STARE show that our method outperforms the state-of-the-art unsupervised methods and achieves competitive performances to supervised methods.

Comments

Q. Liu was supported by the scholarship from China Scholarship Council. B. Zou and Z. Chen were supported by the NSF of China under Grant No.61573380 and No. 61440055. G. Zhao and J. Chen were supported by Academy of Finland, Tekes Fidipro Program and Infotech Oulu.

Rights

Copyright © 2016 the authors

Included in

Ophthalmology Commons

Share

COinS
 
Oct 21st, 12:00 AM Oct 21st, 12:00 AM

Retinal Vessel Segmentation from Simple to Difficult

Athens, Greece

In this paper, we propose two vesselness maps and a simple to difficult learning framework for retinal vessel segmentation which is ground truth free. The first vesselness map is the multiscale centrelineboundary contrast map which is inspired by the appearance of vessels. The other is the difference of diffusion map which measures the difference of the diffused image and the original one. Meanwhile, two existing vesselness maps are generated. Totally, 4 vesselness maps are generated. In each vesselness map, pixels with large vesselness values are regarded as positive samples. Pixels around the positive samples with small vesselness values are regarded as negative samples. Then we learn a strong classifier for the retinal image based on other 3 vesselness maps to determine the pixels with mediocre values in single vesselness map. Finally, pixels with two classifier supports are labelled as vessel pixels. The experimental results on DRIVE and STARE show that our method outperforms the state-of-the-art unsupervised methods and achieves competitive performances to supervised methods.