Refining Self-Supervised Learning in Imaging: Beyond Linear Metric

Along with the development of cloud computing, ensuring fast and secure image exchanges became necessary. in this context, the scientific community has taken a growing interest in JPEG image security. Crypto-compression is generally used to make secure the content of a secret image, while allowing to reduce its size. The drawback of crypto-compression is that it depends on a secret key. indeed, if this key is lost, so is the content of the secret image. in this paper, we propose a joint secret image sharing and JPEG compression method. Shamir's secret sharing scheme is used during the JPEG Huffman coding step, ensuring the visual security of the secret image in the compressed domain, while solving the problem of secret key loss. Our approach is fully format compliant and size preserving when compared to a standard JPEG compression of the original secret image.