Image processing method on radiographic image of piping using MATLAB: Enhancement and detection process

Abstract

Image enhancement is a pre-processing process to enhance the quality and information content of original data. This paper investigates two methods of image augmentation that is deployed to remove noise and improve radiographic images. The first method is image filtering, which consists of smoothing, sharpening and edge enhancement (Sobel & Prewitt) operations. The filtering method emphasizes certain characteristics or eliminates other details. While the second method is morphological technique that utilizes the opening and closing operation, which employed to removed distorted noise and imperfection on the processed images. Each method and operation applied to the image is evaluated subjectively based on the enhance image quality. The image quality measured using MSE (Mean Square Error) and PSNR (Peak Signal to Noise Ratio) which is a full reference metrics. The image quality results are compared to give a wide picture on the performance of the enhanced images. The image processing operations accomplished by using MATLAB image processing toolbox.