Research On Image Denoising Based On Partial Differential Equations

As an important medium, digital image has become an indispensable information source for people to obtain complex information in information age. Limited by imaging methods and conditions as well as the presence of outside interference, the actual image is inevitably contaminated by the noise signal, and difficult to understand, at the same time, having a great impact to the subsequent image processing step. Therefore, improving the performance of the denoising algorithm is very important to improve the performance of the entire image processing system.This article focuses on the image denoising method based on partial differential equations, as well as its application to medical images. The main content of the paper can be summarized as follows:This paper presents details of the partial differential equations applying to image denoising, and analyzes the advantages and disadvantages of some image denoising model based on partial differential equations. the denoising model based total variation and the fourth-order partial differential equations LLT denoising model is studied. By researching their advantages and disadvantages in processing images, a new improved model is proposed. The improved model is also applied to the multiplicative noise filter.The distribution of speckle noise which is existed in medical ultrasound images is introduced, and the corresponding denoising model processing the speckle noise is studied. Based on the adaptive total variation denoising model and the adaptive fourth-order PDE LLT denoising model, a new adaptive PDE denoising model is presented, which can process the multiplicative noise including speckle noise. The method introduced a spatially and scale adaptive function, the improved regularization item is gained. And the fidelity item is constructed via noise distributions. The adaptive model can effectively filter the noise in images, and protect the image edges and texture details at the same time.Finally, improved model of each chapter is applied to filter speckle noise in medical images.The experiments show that the proposed model has obtained good results.