Biometric identification technology has become increasingly common in our daily lives as the demand for information security and security legislation grows around the world. Together, biometrics and encryption can increase the confidence of the lawful data bearer. In cryptographic systems, key generation is an important topic. As a result, how to use biometrics in cryptographic systems (or bio-cryptosystems) is often tied to how to integrate biometrics with cryptographic “keys”. In this paper, since iris recognition is largely considered to be one of the most reliable solutions available we used this metric in order to determine the authorization of persons for using data on the cloud. The Convolution Neural Network (CNN) model for iris images was proposed for the purpose of classifying and identifying authorized and unauthorized persons. Also, the LDA algorithm was used to generate the key from the iris image features. The main idea of key generation is to use it for the purpose of encrypting user data that will be stored in the cloud. In this case, the data of each user will be encrypted depending on the features of his iris. Then, encrypt the generated key using 3DES algorithm. The experimental results of the proposed model show the high accuracy reaching 100% and the encryption efficiency metrics refer to its strength for confronting attacks on data in the cloud.