Abstract—In this paper, a new method for noise reduction in images using partial differential equations (PDEs) has been proposed. In this approach, the method of noise reduction in spatial series using PDEs has been applied in spatial domain image. The image is initially represented as a spatial data matrix and the singular values of this matrix are used to divide the noisy image data into signal subspace and noise subspace. Since singular vectors are the span bases of the matrix, reducing the effect of noise from the singular vectors enhances the information embedded in the matrix. The proposed technique utilizes the PDEs for noise attenuation from the singular vectors. To evaluate the performance of the proposed method, a number of experiments has been performed on both real and synthetic images. The results indicate that the proposed method outperforms the existing approach in image de-noising.