A hidden watermark method with swam particle optimization (PSO) in discrete wavelet waves (DWT) is proposed. The watermark is embedded in a digital image by quantifying the adjacent wavelet coefficients in the wavelet trees and can be extracted blindly. The digital watermark technique incorporates specific information such as the owner identifies, copyright messages or logo images into a digital image. In the watermark, there is a conflict between robustness and imperceptibility. Many methods have been proposed to optimize robustness and imperceptibility. These proposed methods can be roughly classified into two categories. One is the intelligent optimization algorithm (IOA) and the other is the human vision system (HVS). The IOA, like genetic algorithms (AG) and optimization of the swarm of particles (PSO), is to find the most appropriate value of the fitness function to approach the optimal solution. In 2008, Lee et al proposed a hybrid watermarking algorithm based on GA and PSO. In 2011, Wang et al. proposed the PSO algorithm to improve the robustness and imperceptibility of Lin et al.