International Journal of Computer Science Issues

More than a traditional journal...

Retinal vein occlusion (RVO) and diabetic retinal disease is the most widespread type of retinal disorders. The precisely extracted Retinal blood vessel provides a vital factor in the early diagnosis of retinopathy. In this paper a programmed method is proposed to enhance the performance evaluation of feature extraction technique to obtain the affected ocular fundus images in the retinal blood vessel database. The ophthalmologists make use of these tools for patient screening, treatment evaluation, and clinical study. The study consists of two parts: an image acquisition and an adept algorithm for performance rate calculation to detect the retinal vein occlusion in human eye. The features are extracted using completed local binary pattern and the extracted datas are classified using artificial neural network. The precise results exhibit the feasibility of the proposed system in terms of performance evaluation.