An optimized color image watermarking algorithm with stationary wavelet transform and singular value decomposition has been proposed in this work. The algorithm uses two watermarks to address the issues of copyright protection and tamper localization. A gray-scale image watermark is embedded in the transform domain of the green plane of the cover image to address the issue of copyright protection. A binary logo watermark is inserted in the spatial domain of the blue plane of the cover image by the least significant bit approach. It helps to locate the visible and invisible tampers in the cover image. This work is optimized through the genetic algorithm (GA) to achieve better perceptual quality and robustness. A novel fitness function has been proposed to use with the GA to figure out an optimal scaling factor for the embedding process. The strength of the algorithm has been examined by enforcing several attacks on the watermarked image. The embedding capacity of the algorithm and the time complexity of the proposed fitness function are compared with other existing works.

本文提出了一种基于平稳小波变换和奇异值分解的优化彩色图像水印算法。该算法使用两个水印来解决版权保护和篡改定位问题。在封面图像绿色平面的变换域中嵌入灰度图像水印以解决版权保护问题。通过最低有效位方法在封面图像的蓝色平面的空间域中插入二进制标志水印。它有助于定位封面图像中的可见和不可见篡改。这项工作通过遗传算法（GA）进行优化，以实现更好的感知质量和鲁棒性。已经提出了一种新的适应度函数与 GA 一起使用，以找出嵌入过程的最佳缩放因子。该算法的强度已通过对水印图像实施多次攻击来检验。算法的嵌入能力和提出的适应度函数的时间复杂度与其他现有工作进行了比较。