A method of the image Contourlet threshold de-noising based on chaotic particle swarm optimization is proposed. This method can acquire the optimal threshold using chaotic particle swarm optimization in the Contourlet transform domain and then remove the noise by soft threshold function. It does not need the prior information of noise variance. The experimental results show that this method can effectively eliminate the mixed Gaussian white noise and Pepper Salt noise , increase the peak signal to noise ratio(PSNR) and preserve the images details and texture well compared with the de-noising methods of Bayesian wavelet threshold, wavelet threshold by particle swarm optimization and adaptive Contourlet threshold. So the proposed method can improve significantly image visual effect.