Linear filter theory based on Wiener filtering is well understoodand used widely in many fields of image and signal processing. However,the use of linear filters is generally associated with implicitapproximations. Therefore, in this work a series of non-linear filtersis developed based on the concepts of Volterra series and these areapplied to image interpolation problems. More explicitly the aim is tointerpolate one field of a frame of a television picture to form anestimate of the second field. This is known as de-interlacing and isuseful in many areas of video processing, for example standardsconversion. Conventional de-interlacing systems use a fixed linearcombination of the pixels in the aperture. In this paper we consider theextension of these methods to allow estimators based non-linearcombinations of pixel values
展开▼
