BLOOD VESSEL SEGMENTATION USING MINIMUM SPANNING SUPERPIXEL TREE DETECTOR
Blood Vessel Segmentation is of high interest in medical image analysis since the analysis of vessels is crucial for diagnosis, treatment planning and execution, and evaluation of clinical outcomes in different fields, including laryngology, neurosurgery and ophthalmology.Diabetic Retinopathy is a main cause in people and it is a challenging factor in Ophthalmology. The aim of this paper is to segment the vessels in the retina. We are using Minimum Spanning superpixel tree detector method in this paper. It gives slightly better performance in case where minimum weight edge is required from the starting phase of minimum spanning tree formation. The Superpixels provide a more natural and efficient way to compute local image features. It also distinguishes the vessels from the retinal structures. This method consists of three stages. They are Preprocessing, Feature extraction and Segmentation. Preprocessing Step includes contrast enhancement and retinal boundary growth. The contrast limited adaptive histogram equalization (CLAHE) algorithm is used and it generate image with the effects of local contrast enhancement. Then applied a boundary germinating method using iteration based computing to expand the concerned area. Feature extraction step includes illumination Layer, Reflectance layer and texture Layer.