On scanning format and MPEG-2 coding efficiency

The traditional analog video broadcast standard is an interlaced format. In terlace was included in the standards for television broadcasting for very good reason; however, withe the introduction of digital video broadcast (MP EG-2), we should reconsider the use of interlace as part of a digital video chain. Moreover, video format conversation, in general, may also require r econsideration as it can be performed either at the studio end (transmitter side) or at the receiver end (television set). This paper examines the eff iciency of MPEG-2 coding for interlaced and progressive video, and also com pares de-interlacing and picture-rate upconversation before and after codin g. We found receiver side de-interlacing and picture-rate upconversation, i .e., after coding, to give better image quality at a given data rate. Moreo ver, in contrast with other publications, we found interlaced video coding to be better than progressive video for many relevant sequences, even when comparing the results on a progressive display. The paper explains how thes e earlier conclusions were drawn and why the new ones are better.