Retinal diseases become more complicated for the humans. Among this, age-related macular degeneration (ARMD) is the eye-related disease that may cause vision loss. ARMD consists of two types such as dry-ARMD and wet ARMD. The dry ARMD gets affected slowly, so there is no vision loss. The wet ARMD shows the impact on vision loss. If the person got infected for two eyes, they may loss their quality of life. The dry form of age-related macular degeneration tends to get worse slowly, so you can keep most of your vision. The wet form of macular degeneration is a leading cause of permanent vision loss. If it is in both eyes, it can hurt your quality of life. Early detection of ARMD can prevent the vision loss for elderly persons. Deep learning (DL) is an artificial intelligence (AI) technique that works better on human body parts by generating patterns for decision-making. In this paper, discussed several preprocessing techniques, feature extraction techniques and early diagnosis of ARMD diseases by using deep learning algorithms. The performance of various algorithms is discussed on optical coherence tomography (OCT) dry and wet images.