A quantum-inspired post-processing method using collaborative mapping operators for random numbers and its application in image encryption

Random number generators play a crucial role in secure communication, cryptography, and stochastic computation. However, the raw random sequences obtained from physical or algorithmic sources often exhibit statistical bias or correlation, limiting their direct applicability in high secure applications. Drawing inspiration from the projective measurement of polarized photons in the BB84 protocol, this work develops a post-processing method using collaborative mapping operators to enhance the statistical quality of random sequences. The proposed method effectively suppresses the correlation and bias of random sequences through different operators. Experimental results show that the processed sequences achieve excellent statistical properties (NIST SP 800-22 randomness test, autocorrelation, linear complexity, etc.) while maintaining a high throughput of 80 Mbps. Comparative analysis with classical post-processing algorithms such as the Von Neumann method, Toeplitz hashing, and ZCA whitening method demonstrates that our method achieves superior performance in terms of statistical quality, computational efficiency, and implementation simplicity. The successful application of processed sequences in image encryption validates the robustness and practical utility of the proposed post-processing framework.