Application of Kriging for Controlled Minimization of Large Data Sets

摘要

Frequently, the quantity of data available is much greater than that which can bemanipulated in an efficient and timely manner. This can cause several problems. The first, and probably most critical, problem is the excessive on-line storage needs of these huge data sets. Secondly, in the computer animation field, huge data sets may require excessive computational time for generation of each frame of a computer animation. Thirdly, computer screens have a limited resolution and need too much computational time removing excessive detail from images generated with a higher resolution than can be displayed. Lastly, too much time is required to transmit huge amounts of data from location to location. What is needed is a method of minimizing the data set based on some acceptable level of resolution. This thesis develops the application of kriging for the controlled minimization of large data sets based on a maximum acceptable level of error. Specifically, the geostatistical estimation technique of kriging is used to produce minimal data sets and to estimate the unknown values on an arbitrarily sized grid using as input any data set. All the procedures necessary to improve the accuracy of the estimate as well as the kriging procedure are developed. Using these procedures the entire process can be automated. The techniques are demonstrated using Magnetic Resonance Image data to support minimizing on-line storage requirements. Two concurrent thesis efforts also use the techniques to enlarge satellite photographs and to change the grid resolution of terrain data.