This paper implements a hybrid encryption algorithm by combining multiple chaotic systems with the Sea Lion Optimization algorithm (SLO) and Discrete Hartley Transforms. Before the input image is converted to frequency domain, it is confused using Lorenz’s chaotic system and partitioned into four equal parts. After transforming the four sections into the frequency domain, the confusion operations are performed on each section using the chaotic strings generated by the Uruk 4D chaotic system. Using the Sea Lion Optimization (SLO) algorithm, a third phase of confusion is implemented to increase randomness and break down the statistical correlation between the image’s components. The final image is encrypted using a bit-stream from a chaotic Nahrain 3D system. The results of security tests and analyses conducted with the new hybrid algorithm indicate that it provided a high level of resistance to predictable statistical invasion in comparison with other encryption conventional techniques.