Scan-rate conversion
Once the video has been de-interlaced, a scan-rate conversion process may be necessary in order to insure that the input frame rate matches the out- put display refresh rate. In order to equalise the two, fields may need to be dropped or duplicated. As with de-interlacing, some sort of filtering is desirable to smooth out high-frequency artefacts caused by creating abrupt frame transitions.
Image scaling
Flat panels, plasma, LCD or DLP have a fixed format and a fixed resolution to which all video information must adapt. The purpose of the image scaling (or image conversion) chip, normally referred to as the scalar, is to convert the incoming signals which may be NTSC, PAL or computer- generated VGA into the native resolution of the flat panel.
Video scaling is very important because it allows the generation of an output stream whose resolution is different from that of the input format. The only way to avoid image scaling is to crop the image to fit within the confines of a smaller panel. This is the cheapest method but not a very satisfactory one.
Depending on the application, scaling can be done either upwards or downwards. It is important for the scalar to distinguish between the video content and the non-video elements such as text. Failure to do so would distort the non-video parts of the picture making text unreadable or cause some horizontal lines to disappear in the scaled image.
The most straightforward methods of scaling involve either dropping pixels or duplicating existing pixels. That is, when scaling down to a lower resolution, a number of pixels on each line (and/or some number of lines per frame) can be discarded. While this represents a low processing load, the results will yield aliasing and visual artefacts.
A small step upward in complexity uses linear interpolation to improve the image quality. For example, when scaling down an image, interpolation in either the horizontal or vertical directions provides a new output pixel to replace the pixels used in the interpolation process. As with the previous technique, information is still thrown away, so artefacts and aliasing will again be present.
If the image quality is paramount, there are other ways to perform scaling without reducing quality. These methods strive to maintain the high- frequency content of the image consistent with the horizontal and vertical scaling, while reducing the effects of aliasing. For example, assume an image is to be scaled by a factor of Y in the vertical and X in the horizon- tal directions. To accomplish this scaling, the image could be up-sampled (interpolated) by a factor, Y, filtered to eliminate aliasing and then down- sampled (decimated) by a factor X. In practice, these two sampling processes can be combined within a single multi-rate filter.