Here we address the detection of Hemorrhages and microaneurysms in color fundus images. In this proposed work, the input image taken as a fundus image dataset are implemented as input image. The input images are taken in the format .jpg or .png. The preprocessing stage, the collected fundus images are subjected to preprocessing. In the Preprocessing step we can implement the histogram equalization and RGB channel separation are performed. In the feature extraction process, we can implement the (LBP) Local Binary Patteren. LBP is widely used to extract the features from medical images. The feature selection process, we can implement the GA Genetic Algorithm. In this classification process, we can implement the SVM classifier. In machine learning, support vector machines (SVMs, also support vector networks) are supervised learning models with associated learning algorithms that analyze data used for classification and regression analysis.