Master of Science in Computer Science

Faculty / School

Faculty of Computer Sciences (FCS)


Department of Computer Science

Date of Submission



Dr. Muhammad Sarim

Project Type

MSCS Survey Report


There are various methods used in past years for detection of Diabetic Retinopathy. Techniques discussed in this report are founded on Image Processing for identification and Machine Learning for classification for identified object. Different methods of feature extraction in funds images discussed in this paper includes vessel tracking, matched filtering and pixel classification. Whereas for identification and classification of DR based lesions and objects includes Mathematical Morphology, Region Growing, clustering and pixel-based classification. Classification Algorithm mainly used are ANN, FCM and KNN. Above methods are used in detection of multiple abnormalities due to DR which includes Exudates, Microaneurysms (MA) and Hemorrhages (HA).

The full text of this document is only accessible to authorized users.