Volume 32, Issue 1, 2023
DOI: 10.24205/03276716.2023.7002
A Parallel Biogeography Optimization-Based Feature Selection Architecture for Fundus Classification
Abstract
Diabetic Retinopathy (DR) is one of most common eye disease suffered by people with diabetics. The DR can be detected and classified using fundus retina images by employing efficient feature selection techniques. One of the challenges in fundus image classification is the high dimensionality of the features extracted from the images. There are several techniques for classification and feature extraction that are suggested earlier for texture analysis. Feature selection is utilized to find the optimal feature subset. In this work, various metaheuristic methods like Biogeography-based Optimization (BBO) and Particle Swarm Optimization (PSO) are used to find the optimal features. These algorithms are based on the population that explores the space of a certain problem to identify optimal parameters. The Support Vector Machine (SVM) is a machine learning that works well for the problems of binary classification. The proposed methods optimize the feature selection and SVM parameters to enhance image classification.
Keywords
Biogeography-based Optimization (BBO), Particle Swarm Optimization (PSO), Support Vector Machine (SVM), parallel Biogeography-based Optimization and AdaBoost, feature selection, precision, recall