Vessel Border Segmentation of OCT Images Using Textural Correlation

With the development of medical science, optical coherence tomography (OCT) imaging technology is widely used in more and more medical fields, processing of the OCT image has become more important. vascular segmentation of OCT images based on the texture Correlation has been applied to the processing of the image of the microcirculation OCT in vivo, such as retinal tissue, It can help doctors to diagnosis the vascular disease better. In this paper, the main processing steps consist of vascular segmentation and vascular reconstruction. More specially, binary threshold selection based on graylevel expectations and classical Niblack algorithm are used for vascular segmentation. Then, morphological processing is then adopted to build a basic outline of vessels maps. Finally we show the microcirculation maps of the mouse ear can be generated by the proposed scheme.