One of the most promising biometric recognition methods is iris recognition. This is because the iris texture has many features such as freckles, wreaths, stripes, furrows, crypts, etc. These traits are unique and distinguishable to different people. Such unique features in the anatomical structure of the iris make it possible to distinguish between individuals. Therefore, in recent years, many people have tried to improve performance. This article first explains various general steps for the iris recognition system. Then a special type of neural network is used for the recognition part. Experimental results show that high accuracy can be achieved, especially if the first few steps are performed well.