DocumentCode :
3705070
Title :
An image processing based method to identify and grade conjunctivitis infected eye according to its types and intensity
Author :
Joydeep Tamuli;Aishwarya Jain;Aaishwarya V. Dhan;Anupama Bhan; Malay Kishore Dutta
Author_Institution :
Amity School of Engineering and Technology, Amity University, Noida, India
fYear :
2015
Firstpage :
88
Lastpage :
92
Abstract :
Inflammation of the conjunctiva and pain and discomfort in the inner surface of the eyelids is referred to as Conjunctivitis. It causes severe pain, burning sensation or in extreme cases blindness of the eye. Normally conjunctivitis is detected by eye specialist doctors and their limited number makes it difficult for everyone to reach them and get themselves diagnosed. This paper describes an automatic efficient image processing based method to identify conjunctivitis infected eye from a normal eye and classify it according to its types. Some statistical and texture features were used and then followed by PCA for extraction of discriminatory features and then classified using supervised learning method such as multi-class SVM and KNN. The intensity of the infected eyes were also calculated using the significant red plane. Plotconfusion was used to calculate the accuracy and a high accuracy was achieved using this method. Also in addition this proposed method is efficient, computationally fast and costs very low.
Keywords :
"Feature extraction","Support vector machines","Principal component analysis","Entropy","Correlation","Classification algorithms","Microorganisms"
Publisher :
ieee
Conference_Titel :
Contemporary Computing (IC3), 2015 Eighth International Conference on
Print_ISBN :
978-1-4673-7947-2
Type :
conf
DOI :
10.1109/IC3.2015.7346658
Filename :
7346658
Link To Document :
بازگشت