Secure multi-image embedded and encrypted of retinal images via SSCS map

Aiming at the security problems of medical images at present, the paper proposed a retinal image compression and encryption scheme based on chaotic mapping. The retina image is compressed and the compressed information is embedded into multiple color images. This paper adopts different compression methods for different areas to ensure the compression rate. The sensitive area of retina image is identified by Sobel operator edge detection method. Huffman lossless compression is used for sensitive area, and JPEG lossless compression is used for non-sensitive area, which meets the accuracy requirement of medical image and attains good compression efficiency. The Least Significant Bit(LSB) algorithm is combined with chaotic mapping, and the CLSB algorithm is proposed to embed the compressed information into the color image. The embedding position of the algorithm is determined by the random chaotic sequence. Compared with recent research schemes, the CLSB algorithm improves the PSNR of decrypted images by 5.64 %. Correspondingly, to further ensure the security of retinal images, multiple carrier images are jointly encrypted. The proposed Sine-Singer chaotic system(SSCS) has better chaotic characteristics and more complex dynamical behaviors. Experimental results have shown that when applied in encryption algorithms, it exhibits improved robustness and is capable of resisting common attacks such as cropping and salt-and-pepper noise.