论文部分内容阅读
本文介绍一种二维随机模型的生成方法,该模型是一类二维随机域。它是用随机棋盘化平面生成的。这种随机模型提供实际图象的有代表性的典型边沿结构。该模型区别于具有与图象信源匹配二阶特性的二维自回归(auto(?)——gressive)模型(又称AR模型)。在此类模型基础上,可建立一种设计二维DPGM预测器的新方法。它区别于一般的在二维自回归模拟假设上的设计方法,在某种程度上前者更加优越。用这种新方法设计二维DPCM预测器算法。使预测器有能力对付信道误差。文中最后介绍用国产1001电子计算机模拟实现这种随机模型。
This article describes a method of generating a two-dimensional random model, which is a type of two-dimensional random domain. It is generated using a random checkerboard. This stochastic model provides representative, typical edge structures for the actual image. The model is distinguished from a two-dimensional auto-gressive model (also known as AR model) that has second-order matching with the image source. Based on these models, a new method of designing two-dimensional DPGM predictor can be established. It is different from the general design of two-dimensional autoregressive simulation assumptions, to some extent, the former is more superior. This new method is used to design two-dimensional DPCM predictor algorithm. Enable predictor to deal with channel error. Finally, this article describes the realization of this stochastic model with a domestic 1001 computer simulation.