In recent years it has been recognized that embedding information in wavelet transform domain leads to more robust blind watermarks. A major difficulty, however, in watermarking in transform domain lies in the fact that constraints on the allowable distortion at any pixel are specified in the spatial domain. In order to insert an invisible watermark, the current trend has been to model the Human Visual System and specify a masking function which yields the allowable distortion. However, this method is highly suboptimal since it leads to irreversible losses at the embedding stage because the watermark is being adjusted in the spatial domain with no care for the consequences in the transform domain. The central contribution of this work is the proposal of integrated wavelet-based non-linear scaling QIM (NLS-QIM) watermark design to reduce fidelity problem. We consider the watermarking process in conjunction with lossy JPEG2000 compression engine. Experimental results have shown that the proposed watermarking strategy exhibits high robustness to compression attacks with respect to the reconstruction quality. Moreover, it provides a significant capacity improvement compared to other hybrid watermarking coding methods in the literature.