The Retinal hemorrhage detection plays a crucial role in the early diagnosis and treatment of various ocular and systemic diseases, including diabetic retinopathy. This study presents a robust approach for detecting hemorrhages in color fundus images using a combination of preprocessing techniques and rule-based approach. The proposed method incorporates image enhancement, green channel extraction, top-hat filtering, Contrast Limited Adaptive Histogram Equalization (CLAHE) and blood vessel segmentation to improve hemorrhage visibility. Feature extraction is performed to distinguish hemorrhages from other retinal structures, followed by classification Method. The effectiveness of the proposed approach is evaluated on benchmark datasets, demonstrating high accuracy, sensitivity, and specificity. The study analyzes 7,571 hemorrhage blobs extracted from 20 fundus images from the DIARETDB1 dataset. The results indicate that integrating rule-based heuristics with machine learning enhances the reliability of hemorrhage detection, offering a promising solution for automated retinal screening and clinical decision support Keywords—Diabetic Retinopathy, Hemorrhage, Image Processing, Segmentation, Rule-based Classification