Digital images are foremost way of communication in the modern world but the images may possibly noisy due to acquiring process. To ensure the image quality and visibility, image denoising or noise reduction is important process in the image preprocessing techniques that is image restoration. Weld X-radiography images are generally having characteristics such as low-contrast, noisy and small size of feature space. It is necessary to improve the quality of the image for accurate feature extraction. Noise reduction is the challenging task in the denoising process without disturbing of features. An anisotropic diffusion is the famous method for image denoising. This paper studies various anisotropic diffusion based denoising methods and are experimented on noisy weld X-radiography images. Finally, the performance of the methods is evaluated by Mean Square Error (MSE) and Peak Signal-to-Noise Ratio (PSNR) measures. This performance analysis steps forward for further experiment of image segmentation and feature extraction process on noisy weld X-radiography image

anisotropic diffusion, denoise, noise reduction, denoising techniques, weld defect, X-radiography image.