Research On Adaptive Image Denoising Based On Wavelet Transform

In the process of transmission and acquisition,the image is inevitably affected by the noise.For image with less noise,we can recognize the image content according to people's experience.When the image with larger noise interference is needed,we often need professional image processing technology to remove the interference before we can recognize the image.The effective elimination of image noise is an important issue in the field of image processing.The traditional image denoising method is based on the FT denoising method.Because the FT is based on the whole time domain,it is unable to deal with the local feature of the image.The detail features in the image de-noising are removed.In recent years,the denoising method of median filter,mean filter,the denoising method although FT to retain some of the details based on the features of the traditional,but the simulation results show that the median filtering denoising method with non impulse noise image,will cause a loss of details,mean filter denoising method adopts smoothing and edge information of the image will be destroyed.Like low-pass filtering and Wiener filtering,the effect of denoising is different when dealing with different noise types.These methods are less concerned with local features of images,and the details of partial images are difficult to be preserved after denoising.The denoising method based on wavelet transform can make up for the defects of local details missing in previous methods.Wavelet transform is considered from the details of images,while preserving the details of the image while removing the noise.A denoising method based on wavelet transform is adopted.After processing the image to be processed by wavelet transform,the image coefficients and noise coefficients are effectively separated,and then the image after de-noising is obtained after wavelet reconstruction.The denoising methods based on wavelet transform usually have modulus maximum,space domain method and threshold denoising method.The threshold de-noising method based on wavelet transform is the commonly used denoising method,which is divided into soft threshold,hard threshold and improved adaptive threshold method.The simulation results show that the soft thresholds,hard thresholds,and the existing threshold function denoising graphs have the disadvantages of noise residue and blurred image details.Finally,based on the wavelet denoising algorithm based on improved adaptive,the existing threshold selection function was improved after the improved adaptive threshold selection function is more stable in threshold selection,through simulation and experiment,the peak signal-to-noise ratio data and the existing denoising methods are compared,the improved image denoising is more clear and PSNR increase than the existing methods increased by 0.9.In this paper,the improved adaptive threshold method is applied to denoise the plant image.After the denoising image is processed by plant recognition software,the recognition accuracy is higher than the traditional denoising method.