RETINAL VESSELS EXTRACTION BASED ON IMPROVING LINE DETECTION

Abstract

In the medical field, medical images have supported doctors in detecting some dangerous problems related to health. These problems are caused by abnormalities in retinal blood vessels. Therefore, analysis of retinal vascular features is a good way to detect these abnormalities that help doctors make suitable treatment plans in the early stage. One of the retinal analysis tasks is retinal blood vessel extraction which plays an important role because it needs to be done before any measurement can be created. This paper proposed an approach for retinal blood vessel extraction based on multi-scale line detection. In the proposed method, the input image is applied by a multi-scale line detection technique to detect retinal blood vessels. As a result, our method can work effectively to extract more vessels. The performance of the proposed method evaluates both quantitatively and qualitatively on publicly available DRIVE datasets with an accuracy average reaching 96%. The result of the proposed method is better than the other methods.