A GRU and chaos-based novel image encryption approach for transport images

An Intelligent Transport System (ITS) uses smart devices to capture the traffic data in the form of images. However, the adversary can steal and misuse this traffic information. Hence, it becomes essential to have an efficient encryption strategy to save data from various types of attacks. This paper proposes a novel encryption algorithm that uses the Gated Recurrent Unit (GRU) and Sine-Cosine chaotic map to encrypt transport images. The encryption scheme is divided into three phases. In the first phase, two intermediate keys and the seed value required for creating chaotic sequence are generated using unique combinations of 128-bit share key and 128-bit initial vector. In the second phase, permutation is performed using one of the intermediate keys and the chaotic sequence generated by the novel Sine-Cosine chaotic map. The final phase performs the diffusion process using the other intermediate key and GRU approach that uses the chaotic sequence generated by the Sine-Cosine map. The performance of the proposed encryption approach is analyzed using various standard encryption metrics, attacks and decryption parameters. The obtained results and comparative results with existing approaches reveal that the proposed method is suitable for implementing secure and efficient transport image cryptosystems.