Dual-Tamper Detection and Self-Recovery Using Second-order Wavelet Transform for 1080p Images

This paper proposes a dual-tampering detection method for 1080p images, then further recovering its forged content with the recovery information, which is generated by the low-frequency of the second-order wavelet. First, the image is divided into non-overlapping blocks. To improve the quality of the recovered image, we employ a centrosymmetric mapping technique that ensures the security of recovery bits. MD5 is used to generate the first part authentication key based on the most significant bit-plane of the block. SHAI is used to generate the second part authentication key based on the recovery information. By embedding double authentication keys and recovery bits into the original image by replacing the least significant bit, we finally generate the watermarked image. When tampering detection is performed, the extracted authentication keys will be compared with the calculated authentication keys. When tampering detection is performed, the extracted authentication keys will be compared with the calculated authentication keys. If they were not equal, mark the block as tampered. Then use Centrosymmetric mapping to find the second order wavelet low-frequency coefficients for image self-recovery. After the inverse wavelet transform, we use the recovery data to fill into the tampered areas, thus completing the self-recovery of the image. Experimental results show that our scheme has higher invisibility and recovery quality.