Cutset Sampling and Processing

项目摘要切割集取样和处理D.L. Neuhoff（P.I.）该研究项目的重点是开发有效的新方法，用于采样和处理多维数据，如图像，视频或空间分布的传感器数据，特别是第一个和最后一个。传统的数字图像处理方法依赖于获取和处理规则间隔的样本（像素），通常在网格的位置，例如点的正方形网格。相比之下，这个项目正在研究数字图像处理的基础上采取的样本在一个割集，其中最简单的例子是一个正方形网格的线，跨越图像。更一般地说，割集是根据一组离散的潜在样本点之间的邻域关系来定义的。割集采样在图像边缘信息获取方面具有重要意义，其主要动机是二维图像可以用马尔可夫随机场很好地建模，而马尔可夫随机场的未采样像素可以用置信传播算法有效地估计。该研究项目正在研究不同形式的割集采样，并正在开发针对割集采样的重建算法，基于割集的图像压缩方法，用于传感器部署在割集上的分布式传感器网络算法，以及分析技术，以确定割集采样的优点和缺点，并与传统的基于格的采样进行比较。

PROJECT ABSTRACTCutset Sampling and ProcessingD.L. Neuhoff (P.I.)This research project is focused on the development of efficient new methods for sampling andprocessing multidimensional data such as images, video, or spatially distributed sensor data,especially the first and last. Traditional methods of digital image processing rely on the takingand processing of regularly spaced samples (pixels), typically at the sites of a lattice, for examplea square grid of points. In contrast, this project is investigating digital image processingbased on samples taken in a cutset, the simplest example of which is a square grid of lines thatspans the image. More generally a cutset is defined in terms of a set of neighborhood relationsamong a discrete set of potential sample sites. Cutset sampling is expected to be especiallyeffective when the capturing of image edge information is important.The principle motivations for cutset sampling derive from the facts that two-dimensional datacan often be well modeled by Markov random fields and that unsampled pixels of Markovrandom fields can be efficiently estimated with belief propagation algorithms. This researchproject is investigating different forms of cutset sampling, and it is developing reconstructionalgorithms tailored to cutset sampling, image compression methods based on cutsets, distributedsensor network algorithms for scenarios in which sensors are deployed on cutsets, and analysistechniques to determine the strengths and weakness of cutset sampling, and to enable comparisonswith conventional lattice-based sampling.