Locating blood vessels in retinal images using unified Textons

Our research aims to investigate retinal image segmentation approaches based on textons as they provide a compact description of texture that can be learnt from a training set. We propose a new filter bank which is designed for feature extraction. The K-means clustering algorithm is adopted for texton generation. We back project textons onto the ground truth in a machine learning stage to identify the corresponding vessel textons and these are subsequently used to identify vessels in the test set. To verify that these unified textons provide a general tool that can be used for vessel segmentation, we perform experiments on three different data sets, generating textons from the one set, and testing on another two data sets. Our results show the method outperforms other published work, and reveal that it is possible to train unified textons for retinal vessel segmentation.

[1]  Roberto Marcondes Cesar Junior,et al.  Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification , 2005, IEEE Transactions on Medical Imaging.

[2]  J. A. Sethian,et al.  Fast Marching Methods , 1999, SIAM Rev..

[3]  T. Miller,et al.  Blood Vessel Segmentation in Retinal Images , 2004 .

[4]  Lei Zhang,et al.  Retinal vessel extraction by matched filter with first-order derivative of Gaussian , 2010, Comput. Biol. Medicine.

[5]  B. Julesz.  Textons, the elements of texture perception, and their interactions , 1981, Nature.

[6]  Andrew Zisserman,et al.  Unifying statistical texture classification frameworks , 2004, Image Vis. Comput..

[7]  Jitendra Malik,et al.  Representing and Recognizing the Visual Appearance of Materials using Three-dimensional Textons , 2001, International Journal of Computer Vision.

[8]  Ana Maria Mendonça,et al.  Segmentation of retinal blood vessels by combining the detection of centerlines and morphological reconstruction , 2006, IEEE Transactions on Medical Imaging.

[9]  M. Goldbaum,et al.  Detection of blood vessels in retinal images using two-dimensional matched filters. , 1989, IEEE transactions on medical imaging.

[10]  Xiaohong W. Gao,et al.  A method of vessel tracking for vessel diameter measurement on retinal images , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[11]  Max A. Viergever,et al.  Ridge-based vessel segmentation in color images of the retina , 2004, IEEE Transactions on Medical Imaging.

[12]  Hong Shen,et al.  Optimal scheduling of tracing computations for real-time vascular landmark extraction from retinal fundus images , 2001, IEEE Transactions on Information Technology in Biomedicine.

[13]  Shankar M. Krishnan,et al.  Detection and measurement of retinal vessels in fundus images using amplitude modified second-order Gaussian filter , 2002, IEEE Transactions on Biomedical Engineering.

[14]  Frédéric Zana,et al.  Segmentation of vessel-like patterns using mathematical morphology and curvature evaluation , 2001, IEEE Trans. Image Process..

[15]  José Manuel Bravo,et al.  A New Supervised Method for Blood Vessel Segmentation in Retinal Images by Using Gray-Level and Moment Invariants-Based Features , 2011, IEEE Transactions on Medical Imaging.

[16]  A.D. Hoover,et al.  Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response , 2000, IEEE Transactions on Medical Imaging.

[17]  Paul Mitchell,et al.  Retinal Vessel Diameters and Obesity: A Population‐Based Study in Older Persons , 2006, Obesity.

[18]  Paul Mitchell,et al.  Retinal vessel diameter and open-angle glaucoma: the Blue Mountains Eye Study. , 2005, Ophthalmology.

[19]  Donald A Adjeroh,et al.  Texton-based segmentation of retinal vessels. , 2007, Journal of the Optical Society of America. A, Optics, image science, and vision.