Paper Details

A novel splat feature classification method is presented with application to retinal hemorrhage detection infundus images. Reliable detection of retinal hemorrhages is important in the development of automated screening systemswhich can be translated into practice. Under supervised approach, retinal color images are partitioned intononoverlapping segments covering the entire image. Each segment, i.e., splat, contains pixels with similar color andspatial location. An optimal subset of splat features is selected by a filter approach followed by a wrapper approach.Aclassifier is trained with splat-based expert annotations and evaluated on the publicly available Messidor dataset. An areaunder the receiver operating characteristic curve of 0.96 is achieved at the splat level and 0.87 at the image level. Whilewe are focused on retinal hemorrhage detection, our approach has potential to be applied to other object detection tasks.Neural network is proposed for classification in this paper. In many image-processing applications, the attributes ofshapes within images must be extracted and classified.

Anjana K, “RETINAL HEMORRHAGE DETECTION USING NEURAL NETWORK WITH SPLAT FEATURE CLASSIFICATION”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 3, Issue 12, pp. 366-369, December 2016.