Alphatech Inc., Burlington, MA.
IEEE Trans Image Process. 1995;4(2):194-207. doi: 10.1109/83.342185.
A class of multiscale stochastic models based on scale-recursive dynamics on trees has previously been introduced. Theoretical and experimental results have shown that these models provide an extremely rich framework for representing both processes which are intrinsically multiscale, e.g., 1/f processes, as well as 1D Markov processes and 2D Markov random fields. Moreover, efficient optimal estimation algorithms have been developed for these models by exploiting their scale-recursive structure. The authors exploit this structure in order to develop a computationally efficient and parallelizable algorithm for likelihood calculation. They illustrate one possible application to texture discrimination and demonstrate that likelihood-based methods using the algorithm achieve performance comparable to that of Gaussian Markov random field based techniques, which in general are prohibitively complex computationally.
先前已经引入了一类基于树状递归动力学的多尺度随机模型。理论和实验结果表明，这些模型为表示本质上是多尺度的过程提供了一个极其丰富的框架，例如 1/f 过程，以及 1D 马尔可夫过程和 2D 马尔可夫随机场。此外，还通过利用其递归结构为这些模型开发了高效的最优估计算法。作者利用这种结构开发了一种用于计算似然的计算效率高且可并行化的算法。他们举例说明了一种可能的纹理判别应用，并证明了使用该算法的基于似然的方法可以达到与基于高斯马尔可夫随机场技术相当的性能，而后者在计算上通常过于复杂。
