Blind enhancement for ultrasonic C-scans using recursive 2-D H-infinity-based state estimation and filtering

The main purpose of any image enhancement scheme is to improve the quality of image in such a way that more information is available for the observer. Ultrasonic C-scans, which are commonly used in non-destructive evaluation of material, are a special form of images that suffer from blurring effects due to the quantisation in the scanning grid, squaring and windowing opera tions, and additive noise. A software approach is presented in this work wh ich utilises the 2-D H-infinity Estimation and Filtering theories to enhanc e the ultrasonic C-scans. The states in the mathematical formulation are se lected as the neighbouring pixels in an image and the blur components as th e system matrix coefficients. Repetitive application of estimator and filte r on successive images results in an overall improved image.