• Dr Pavithra M
• Kanneti Jaya Pranay
• Lakkireddy Venkatesh
• Anumolu Manasa

“CLEAR VISION GIVES YOU A CLEAR LIFE”, this wording means a lot. Every person wishes to have a good vision to survive, enjoy, and have a peaceful life and most people do the same. A promising good vision is presented by a good retina and its retinal health. As a coin has two sides some diseases affect the retinal layer in our eyes. A healthy retina is essential for clear vision. Retinal diseases are common as they can affect any part of the eye. As retinal diseases affect the vital eye tissue, they cause serious problems and affect vision, some even leading to blindness. Ones if people are affected with a retinal disease they come to know after a severe effect on vision. The sad part is that we don’t even get any pain or wound when we get retinal disease, we can’t predict the symptoms easily. After getting the decrease in vision rate common people will approach the eye hospitals and consult an ophthalmologist. They usually went through a screening test handling with CAD machines, which means a Computer-Aided Diagnosis technique connected with the computer and to the internet connection. which is a complete Artificial Intelligent machine. which will be costlier and common people don’t get a better guideline to approach at the beginning stage. To overcome this issue, we have planned to develop an early predictor of pre-clinical signs of retinal disease symptoms by involving the IMAGE PROCESSING technique. Now we are planning to create a prototype of an “Early predictor of retinal diseases by image processing” and to implement it as a working model of the vending concept and to keep it in public places and allow common people to utilize it with less cost and with minimum handling. We have analyzed so many algorithms to implement our idea, finally, we are satisfied with HAAR CASCADE ALGORITMH which is a MACHINE LEARNING TECHNIQUE because one of the primary benefits of Haar cascades is that they are just so fast — it’s hard to beat their speed. Haar algorithm does not undergo a dual segmentation process for further classification whereas other image processing algorithms will lie under with segmentation and without segmentation processing and it will consume more time. Some other ML algorithms will demand a fundus image for image processing later it will convert the color image to a grayscale image using a scanner, but the HAAR Algorithm does not demand it For all these reasons we have chosen HAAR ALGORITHM and it also satisfies our needs to the extend. We are using OpenCV which is an open-source computer vision for implementing the programming which will be developed in Python language. We have chosen OpenCV because it is platform-independent and it already has predefined library files for image processing which will reduce our working space. We are using RASPBERRY PI as a SOC, it will swap the work of a main frame computer because it is low cost, has huge processing power in a compact board, and has many interfaces. We are using PI CAMERA for image capture resolution with 1080p at 25 frame PI CAMERA has a better graphic processing capability than others.

PREDICTION OF RETINAL DISEASES USING IMAGE PROCESSING TECHNIQUES

"PREDICTION OF RETINAL DISEASES USING IMAGE PROCESSING TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.e705-e714, April-2024, Available :http://www.jetir.org/papers/JETIR2404477.pdf

"PREDICTION OF RETINAL DISEASES USING IMAGE PROCESSING TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppe705-e714, April-2024, Available at : http://www.jetir.org/papers/JETIR2404477.pdf