| Title |
Machine learning-based retinal image processing for detection possibility of blindness / |
| Translation of Title |
Mašininiu mokymusi grįsta akies dugno vaizdų analizė galimam apakimui nustatyti. |
| Authors |
Sarac, Mehmet Furkan |
| Full Text |
|
| Pages |
74 |
| Keywords [eng] |
Image Classification, Image Preprocessing, Color Spaces, ResNet, EfficientNet, Diabetic Retinopathy, Deep Learning, Machine Learning. |
| Abstract [eng] |
The purpose of the thesis is to provide a different point of view to perform diabetic retinal image classification problem and the influences of various preprocessing methods on deep learning methods to solve the problem of the detection of diabetic retinopathy. To experiment the preprocessing methods such as HSV, HSL, and Lab color spaces, EfficientNet and ResNet deep learning architectures are used. Conclusions and detailed experimental results are provided at the end of the research. |
| Dissertation Institution |
Vilniaus universitetas. |
| Type |
Master thesis |
| Language |
English |
| Publication date |
2022 |
