Image Denoising Method with Improved Threshold Function

Since remote sensing image are created uniquely, noise is invariably present. This can degrade the image quality and hinder further processing of the picture. For this reason, image denoising is a crucial step in the image processing process. The limitations of the traditional threshold functions in practical applications, as well as the fact that the threshold functions in previous studies were not continuous and did not discuss the type of noise. Therefore, a new adjustable continuous threshold function is constructed based on existing threshold functions and is used for denoising images with different kinds of noise. Simulation experiments demonstrate that wavelet threshold denoising using the new threshold function outperforms current threshold functions and methods for images with salt and pepper, Gaussian, and speckle noise. Moreover, it effectively filters out noise while retaining more detail. This provides a feasible method for wavelet threshold denoising and can be applied to images containing different noises.