The process of going to an optometrist, doing lengthy eye check examinations and finding the power of prescription lenses is lengthy. Can this process be automated? Can we make a good prediction of the power of the prescription lenses if we have information about the eyes through a photo? Our project addresses these challenges.
What it does
We built an app that uses Google Cloud Vision API, Android Studio and machine learning techniques that provides the power of the prescription lenses to correct the eye defects.It measures features of the eyes and provides a prediction of the power of the lens required to correct the eye defects.
How we built it
We used the Google Cloud Vision API to detect the eyes of a person in the image and calculated the axial radius of the eyes. We had a database of the relationship between axial radius and the power of the lens required to correct the eye defect - myopia or hypermetropia. We then used linear regression to find the power of the lens required to correct the vision based on the photo taken.
Challenges we ran into
Finding the features of the eye that correspond to the eye defects was challenging as we had to do a literature survey of ophthalmology research around this problem.
Accomplishments that we're proud of
What we learned
What's next for EyeCular