automatic detection of multiple oriented blood vessels in retinal images自动检测多个面向视网膜血管图像.pdfVIP

J. Biomedical Science and Engineering, 2010, 3, 101-107 JBiSE doi:10.4236/jbise.2010.31015 Published Online January 2010 (http://www.SciRP.org/journal/jbise/). Automatic detection of multiple oriented blood vessels in retinal images P. C. Siddalingaswamy1, K. Gopalakrishna Prabhu2 1Department of Computer Science Engineering, Manipal Institute of Technology, Manipal, India; 2Department of Biomedical Engineering, Manipal Institute of Technology, Manipal, India. Email: pcs.swamy@ Received 9 September 2009; revised 16 October 2009; accepted 19 October 2009. ABSTRACT be examined by ophthalmologists. Manual diagnosis is usually performed by analyzing the images from a pa- Automatic segmentation of the vasculature in retinal tient, as not all images show signs of diabetic retinopathy, images is important in the detection of diabetic reti- it increases the time and decreases the efficiency of nopathy that affects the morphology of the blood ves- ophthalmologists. Therefore, an automatic segmentation sel tree. In this paper, a hybrid method for efficient of the vasculature could save workload of the ophthal- segmentation of multiple oriented blood vessels in mologists and may assist in characterizing the detected colour retinal images is proposed. Initially, the ap- lesions and to identify false positives [2]. Another im- pearance of the blood vessels are enhanced and ba