An improved multi-objective particle swarm optimization(IMOPSO)algorithm is proposed according to the sorting without domination and crowding distance calculation of the particle individual. The IMOPSO is applied to the design of the digi-tal watermarking system in discrete cosine transform(DCT)domain to propose the digital watermarking algorithm based on multi-objective particle swarm optimization. The algorithm takes the location embedded with the watermarking as the solution of the particle,and the peak signal-to-noise ratio(PSNR)and normalized correlation(NC)as the two optimization objects to get the optimal solution set of the digital watermarking system. The experimental results indicate that the algorithm can get the multiple watermarking embedded schemes to balance the PSNR and NC. In comparison with other related algorithms under the same con-dition,the proposed algorithm can satisfy the watermarking invisibility,and has good robustness aiming at the noise,compres-sion,cutting and other attacks.%根据对粒子个体的非支配排序和计算拥挤距离,提出一种改进的多目标粒子群优化(IMOPSO)算法.将IMOP-SO应用于DCT域数字水印系统设计,提出基于多目标粒子群优化的数字水印算法.该算法以水印嵌入位置作为粒子的解,以不可见性(PSNR)和鲁棒性(NC)作为两个优化目标,得到数字水印系统的最优解集.实验结果表明,提出的算法得到多种满足PSNR值和NC值平衡的水印嵌入方案.与其他相关算法在同等条件下相比较,算法满足水印不可见性的同时,在噪声、压缩、裁剪等攻击下具有较好的鲁棒性.
展开▼
