Abstract: In response to the strong correlation between the chaotic system state and initial state and parameters in traditional chaotic encryption algorithms, which may lead to the periodicity of chaotic sequences, the Chaos-LSTM model is constructed by combining chaotic systems and LSTM neural networks. The chaos sequence proliferation (CSP) algorithm is constructed to address the problem that the limited computational accuracy of computers can lead to the periodicity of long chaotic sequences, making it unsuitable for encrypting objects with large amounts of data. Combining the above two, the geographic information encryption system based on Chaos-LSTM and CSP is proposed. Firstly, the Chaos-LSTM model is used to output chaotic sequences with high Spectral Entropy (SE) complexity. Then, a shorter chaotic sequence is selected and proliferated using the CSP algorithm to generate chaotic proliferation sequences that match the encrypted object, and a randomness analysis is conducted and testing is performed on it. Finally, using geographic images as encryption objects, the chaos proliferation sequence, scrambling, and diffusion algorithm are combined to form the encryption system, which is implemented in the ZYNQ platform. The system’s excellent confidentiality performance and scalability are proved by software testing and hardware experiments, which can be used for the confidentiality peers of various encryption objects with outstanding application value.