An Enhanced Informed Watermarking Scheme Using the Posterior Hidden Markov Model

Figure 5

Average KLD between Θ^attack_post and Θ^attack_pri under different predefined robustness settings (a) for Rbst_QF=70 (b) for Rbst_QF=80 (c) for Rbst_QF=90.

(a)

(b)

(c)