<p>Nowadays, the security of digital data and images transmitted over the internet is critical. This study proposes a novel algorithm for image encryption based on a two-dimensional Tinkerbell chaotic function. One of the merits of using chaotic functions is their ability to generate pseudo-random numbers for image encryption. However, in most chaotic-based encryption algorithms, a similar chaotic sequence is produced for any type of image without considering its nature. In this research, a creative method is proposed based on Particle Swarm Optimization (PSO) to tune the parameters of the Tinkerbell chaotic function. PSO adjusts the Tinkerbell function parameters in such a way that the pseudo-random sequence is compatible with the type of input plain-image, thereby maximizing the entropy rate in the encrypted image. The proposed method has significant practical applications, including the secure transmission of confidential medical images, protection of data in cloud storage systems and real-time communication platforms such as instant messengers. The performance analysis of the proposed algorithm demonstrates its resistance against common attacks in the field of image encryption.</p>