A Detection of Amblyopia using Image Processing and Machine Learning Techniques

Abstract

Amblyopia is an eye disease occurred due to the failure of interconnection between the brain and the eye. It typically affects the vision of children and kids. Using image processing and machine learning techniques validate the user name and password then proceed to capture image and if it is valid then use the canny edge detection techniques to detect the amblyopia. If validation is success means test the training data using KNN Algorithm, Logistic Regression and Random Forest Classifier(RFC). Comparing to other algorithm RFC is more accurate to detect the amblyopia. It is nothing but lack of coordination between eye and brain. It could be a new visual process disorder which ends within the patient perceiving a blurred image from one in all their eyes that isn't rectifiable with glasses or contact lenses.

We used SPYDER IDE, Open CV, and python programming language for image processing and machine learning algorithm.It can result from any condition that prevents from the eye from focusing clearly. The Input parameters taken as for the sample dataset are namely gender, age, cataract, myopia, hyperopia, strabismus, and class.