作者: J. A. Stuller , A. N. Netravali , J. D. Robbins
DOI: 10.1002/J.1538-7305.1980.TB03358.X
关键词:
摘要: This paper presents algorithms for predicting luminance changes in successive television frames. The can result when objects a TV scene move or illumination varies. By gradient search technique, which seeks to minimize functional of the interframe prediction error, we estimate two parameters associated with these — displacement and gain. Using estimates parameters, also develop, coding, adaptive predictors segmentor determine pels need be transmitted. We describe several coder variations compare them by computer simulations using three substantially different sequences. For sequences, gain compensation improved segmentation reduced bit rate conditional replenishment encoder 50.7, 11.1, 39.3 percent. Displacement 61.0, 24.8, 14.5 Combined 63.4, 32.2, 44.6