Title: Morphological statistical features for automatic segmentation of blood vessel structure in retinal images
Authors: Sangita Bharkad
Addresses: Department of Electronics and Telecommunication Engineering, Government College of Engineering, Aurangabad, 431001, Maharashtra, India
Abstract: This paper presents a new method for automatic segmentation of blood vessels in retinal images. The proposed method is based on morphological high pass filter called top hat transform and statistical features. Directional information of vasculature structure is captured using top hat transform with line structuring element rotated at 18 different directions. Statistical features namely mean, median, maximum and variance are used to reconstruct the enhanced vasculature structure. The proposed method obtained four enhanced vasculature structure based on these statistical features. Best statistical feature is determined by evaluating the segmentation performance in terms of accuracy, sensitivity and specificity. This method is evaluated on the publicly available DRIVE database. Segmentation results show that the method outperforms all evaluated segmentation approaches. Its accuracy and fast execution make it appropriate for automatic screening system of retina related diseases such as diabetic retinopathy.
Keywords: blood vessel; segmentation; morphological filter; statistical features.
DOI: 10.1504/IJTMCP.2017.087870
International Journal of Telemedicine and Clinical Practices, 2017 Vol.2 No.3, pp.197 - 214
Received: 13 May 2016
Accepted: 14 Oct 2016
Published online: 06 Nov 2017 *