Research Of Image Denoising Method Based On Partial Differential Equations

Image has always been an effective carrier and preferred way of transmitting information and plays an irreplaceable role in daily life.In order to obtain high-quality images,researchers have adopted many methods.This paper mainly studies the application of partial differential equations in image denoising.The defects of P-M model,CLMC model and Y-K model are mainly improved.The specific research contents are as follows:(1)By analyzing the principle of diffusion function in the P-M model,a new diffusion function is constructed and applied to the regularized P-M model.The experimental results show that this method is more effective than the original method,and has achieved good results in terms of processing speed of images,de-noising effect,and maintaining image characteristics;(2)Combining the basis of the previous part,the P-M model and algorithm of partial differential equations are improved.Based on the regularized model,the fidelity term is added to the model,and the four-domain diffusion of the original algorithm is extended to the eight-domain diffusion.The improved method is superior to the traditional method in all evaluation indicators of overall visual effects and image processing effects,reducing the number of iterations and the generation of "block effects".In contrast,it enriches the detail information of the de-noising image;(3)Aimed at the disadvantages of You-Kaveh partial differential equation model,to improve the model,combining with the former backward diffusion function to propose new diffusion function,use wavelet transform to improve the model algorithm,low threshold of the Canny edge detection operator is used to divide the edge region and edge character region,the diffusion coefficient in the flat area of the image is affected by enhanced diffusion,promote smooth effect,in the plain area and in the region of the image characteristics of image texture detail place to protect it,the results also show that the proposed model can well remove the gaussian noise and maintain image characteristics,Also has the great elimination ability to the salt and pepper noise.The visual effect of the denoised image is improved and the experimental data prove the effectiveness of the new method.