<p>With the rapid advancement of big data, images are increasingly being utilised as a direct means of conveying information. Ensuring that information security remains safe from compromise during image transmission operations has become an urgent and vital necessity. In this paper, a scheme that combines multi-image encryption and LSB steganography based on a memristor neural map is proposed. A first-order memristor model is established and coupled with a Rulkov neuron to form a new chaotic map, called SOM-Rulkov. Through analysis of its phase diagram, Lyapunov exponents, bifurcation diagram, and complexity metrics, demonstrates that the SOM-Rulkov map exhibits complex dynamical behaviour. DSP technology enables the implementation of complex nonlinear systems on digital platforms, and the SOM-Rulkov map can be achieved on a DSP. Iterations of the SOM-Rulkov map can generate chaotic sequences, while steganography techniques and encryption algorithms are incorporated into the scheme. First, secret images of different types and sizes are embedded into the carrier image by LSB technology, ensuring the secret information remains unnoticed, then the carrier images containing the secret images undergoes confusion and diffusion to form cipher images. Finally, simulation results demonstrate that this scheme exhibits robust performance, a larger key space, strong key sensitivity, and resilience against differential attacks and statistical attacks, thereby offering enhanced security, while also offering significant efficiency advantages.</p>