Analysis of the “Wavelet Tree Quantization” watermarking strategy and a modified robust scheme

Abstract

Here we study a recently proposed watermarking scheme based on the paper "Wavelet Tree Quantization" (WTQ) by Wang and Lin (IEEE Trans Image Process 13(2): [154][155][156][157][158][159][160][161][162][163][164][165] 2004). In given scheme, wavelet coefficients corresponding to the same spatial locations are grouped together. Two such groups, selected at random, constitute a supertree. Some of these supertrees are quantized to embed the watermark information in the image. In the process of cryptanalysis we first identify the groups which are quantized during the watermark insertion process. Then we select the nonquantized groups and quantize them too, to remove the watermark. Experimental results show that the watermark is completely removed by this attack. The cryptanalysis falls under the cipher text only jamming attack which requires only a single watermarked copy. Further, This is an extended and revised version of the paper we suitably modify the WTQ scheme to make it robust against such cryptanalytic attacks.

Tree group based Wavelet Watermarking using Energy Modulation and Consistency Check (WW-EMCC) for digital images

“…The difference between WTQ, MWTQ and WTGM is that the super tree selection in WTQ is random but the selection in MWTQ and WTGM is based on the sorting of the energy summation through the trees. Therefore, MWTQ and WTGM need the record of the super tree ordering [11,30] which also has the same maximum number of embedding watermark bits. In summary, the most possible watermark capacity is 768 for WW-EMCC which is equivalent to the maximum number of embedding watermark bits for WTQ [33], MWTQ [11] and WTGM [30] algorithms.…”

“…Even if the attacker has no knowledge of which two trees are used for embedding, he can still quantize those super trees that are not quantized earlier with respect to the estimated quantization indices. Das and Maitra had presented how this could be accomplished in [11] by using cryptanalysis approach to attack WTQ. Since the weakness of WTQ could not provide the security promise of watermarking, Das and Maitra [11] had proposed a modified WTQ (MWTQ) algorithm which essentially used the positive and negative modulation (bilateral modulation) [22] to embed the watermark instead of quantization to defy the cryptanalysis attack.…”

“…Das and Maitra had presented how this could be accomplished in [11] by using cryptanalysis approach to attack WTQ. Since the weakness of WTQ could not provide the security promise of watermarking, Das and Maitra [11] had proposed a modified WTQ (MWTQ) algorithm which essentially used the positive and negative modulation (bilateral modulation) [22] to embed the watermark instead of quantization to defy the cryptanalysis attack. In general, the MWTQ scheme is similar to the wavelet tree group modulation (WTGM) [29,30] technique which not only adopts the sum-of-subsets strategy [21] to efficiently group the wavelet trees but also further uses the contrast sensitive function (CSF) [17] and noise visibility function (NVF) [32] of human visual system for the better visual quality of the watermarked image.…”

“…[2,10,11,21,22,24,29,30,33]. Spatial domain watermarking embeds information by directly modifying the value of image pixels, e.g., replacing the least significant bit (LSB) of image pixels with a binary pseudorandom noise (PN) sequence as watermark information.…”

“…Because frequency-domain watermarking schemes tend to achieve both perceptual transparency and robustness requirements, various related algorithms have been developed [2,8,10,11,21,22,24,29,30,33]. …”