Research On Image Denoising Method Based On Adaptive Diffusion Filtering

In the process of acquisition,transmission and storage of image information,various kinds of noise will be mixed into the image for various reasons.Therefore,before acquiring image information,removing noise from the image and improving the quality of the image are important parts of image processing technology.how to solve the contradiction between effectively removing noise and preserving important details has become a hot spot in image denoising research.In this paper,the classical partial differential equation denoising model is firstly studied and analyzed,and then it is improved,and a new denoising method is proposed.Finally,the conclusion is drawn through simulation experiments.The main work of this paper is as follows:1.Because both the traditional algorithm PM and YK model has shortcomings in image processing,we put forward a new mixed model.As we have already know the relationship between gradient and threshold,We put the pending image into different districts,different models will be used in different districts,and we will use Additive Operator Splitting(AOS)algorithm to solve the model.Thus,the model can effectively remove the noise,and maintain the detail of image edge and texture as well.The experimental results show that compared with the PM model and the YK model,the proposed model has better denoising performance.2.On the basis of PM model,the diffusion coefficient in diffusion equation is improved,and the fitting diffusion coefficient is established to overcome the shortcomings of loss of texture detail information and edge degradation caused by excessive intensity.Then,the threshold function is designed adaptively to automatically control the threshold according to the maximum gray value of the image and the number of iterations,thereby further retaining important detail features such as image edges and textures.Finally,the designed algorithm is simulated.The experimental results show that the peak signal-to-noise ratio(PSNR)of the new algorithm is greatly improved compared with the classical algorithm,and the image edge and detail information are protected while effectively suppressing noise.3.Firstly,starting from the intrinsic orthogonal coordinate system of TV flow diffusion equation,a simplified orthogonal structure is proposed and an orthogonal diffusion filtering model is established to solve the problem of “blocky” effect in the denoising process.In order to protect the detail information such as edge texture,the orthogonal structure is further simplified,and compared with the classical denoising model,an adaptive orthogonaldiffusion filtering model is proposed.The new model can adaptively denoise and protect the edge texture and other details.Compared with the orthogonal diffusion filtering model,the flexibility is high,and the diffusion coefficient can be adjusted according to different regions of the processed image,the degree of smoothness can be controlled,and a clear image can be processed more reasonably.The simulation experiments show that the proposed image denoising model has better performance,the peak signal-to-noise ratio is greatly improved,and the detail information such as edge texture remains intact.