Efficient Cipher-Image Coding via Compressive Sensing and Auxiliary-Information-Guided Mapping for Secure Cloud Storage in Consumer Electronics

Storing images in the cloud effectively alleviates the local storage burden on typical consumer electronic devices such as smartphones and tablets. However, to safeguard privacy, consumers increasingly adopt encryption as a preemptive measure prior to uploading, which gives rise to an urgent requirement for efficient encoding of encrypted images. Recently, cipher-image coding by compressive sensing (CS) has attracted more and more attention. However, existing schemes fail to deliver satisfactory compression performance, primarily due to limitations in the measurement coding and reconstruction processes. To address these challenges, we propose a cipher-image coding scheme based on CS and auxiliary-information-guided mapping (AIGM) for secure cloud storage in consumer electronics. First, we design an AIGM operation for prepossessing the original image. This AIGM operation effectively reduces the range of the CS measurements, significantly saving the number of bits required for encoding the CS measurements. Second, leveraging alternating direction method of multipliers (ADMM), we propose an iterative denoising-enhanced bivariate shrinkage (IDBS) algorithm. This algorithm integrates a denoising operation with a bivariate shrinkage operation during the iterative reconstruction process. It can substantially enhance the quality of the reconstructed image, which is also beneficial for ratio-distortion (R-D) performance improvement. Experimental results demonstrate that our scheme outperforms the state-of-the-art CS-based schemes in terms of R-D performance. More notably, the proposed scheme applied to encrypted images demonstrates superior R-D performance compared to JPEG, which uses original images as inputs. In contrast, existing CS-based schemes impose a substantial penalty on R-D performance.