Glaucoma is a progressive optic neuropathy that causes irreversible vision loss and is considered one of the leading causes of blindness worldwide. Early diagnosis of glaucoma is essential to prevent permanent damage to the optic nerve and preserve visual function. However, manual examination of retinal fundus images is time-consuming, subjective, and highly dependent on expert ophthalmologists. Consequently, automated glaucoma detection systems based on image processing, machine learning, and deep learning have gained significant attention in recent years. This survey paper presents a comprehensive review of existing glaucoma detection methodologies using retinal fundus images. The study covers traditional image processing approaches, wavelet-based feature extraction techniques, decomposition models, texture and statistical feature analysis, machine learning classifiers, and recent deep learning frameworks. Furthermore, the paper discusses preprocessing methods, feature fusion strategies, dimensionality reduction techniques, and classification models used for glaucoma diagnosis. A comparative analysis of existing techniques is also presented to identify their advantages, limitations, and research gaps. The review highlights that hybrid frameworks combining multi-resolution analysis and machine learning provide an effective balance between accuracy and computational efficiency. Finally, future research directions toward robust, efficient, and clinically deployable glaucoma detection systems are discussed.

Keywords: Glaucoma Detection, Fundus Images, Machine Learning, Deep Learning, Discrete Wavelet Transform, Feature Extraction, Medical Image Processing, Retinal Image Analysis.