Research on Data Compression Algorithm for Isoline

With the development of contemporary science and technology, computer has been used to imitate geological physiognomy.With volumes of sample points, we may produce simulacrum that can reflect real geographical information from original figure. However the redundancy data decrease velocity and cut down memory of the computer enormously.Base on above reasons, data compression was brought to a significant place. The data compression of isoline comes along with the development of the GIS, particularly WebGIS. Hitherto, in Geology Statistics, many mature algorithms have been presented such as DP etc. Review the numerous data compression algorithms, the essence of all is a process of electing characteristic points, in other words, they are a process of elimination for the redundant points. The paper combines with the relevant theory of DP algorithm and present a picecwise DP algorithm that is base on isoline. Moreover on the basis of the spline interpolation, the points can be further reduced by means of electing the extreme points. Finally it is will be proved by experiment that this algorithm is effectual for the compression of isoline interpolated with cubic parameter spline. This paper is organized as follow: Chapter 1: Interpolation and fitting from scatter points.Introducing some interpolation algorithms in the geology statistics that are usually used including of Moving Average Method, Krige etc.Because of demands, the paper introduces cubic spline interpolation with great importance finally. Chapter 2: Data compression of curve.Giving a classification to existing compression algorithms, and introducing some compression algorithms the predecessors have advanced with pictures. Chapter 3: Data compression of isoline.First introducing a algorithm that is applicable to cubic spline curve.Secondly presenting picecwise DP compression algorithm.Finally achieving the data compression for isoline with above two algorithms and giving some examples.

Alternate abstract:

随着现代科学技术的快速发展，用计算机来模拟地质地貌已成为可能。通过采集大量的样本点信息点，我们可以构造出能真实反映地物原貌的图形。但大量的数据在存储和处理速度上都大大降低了计算机的工作效率，从而数据的压缩处理就被提到了一个重要的位置。 等值线数据压缩是随着地理信息系统（Geographical Information System，GIS）的发展，特别是WebGIS的普及而产生的。在地质统计学方面，目前较成熟的方法有垂距法、DP算法等。纵观前人给出的众多数据压缩方法，其实质都是对保形特征点的一个提取保留的过程，或者说是对冗余数据点的剔除过程。本文结合DP算法的有关理论，给出了一种基于等值线图的分段数据压缩算法，并且根据样条插值的相关特点，通过提取极值点来实现进一步压缩的目的。最后通过试验发现，该算法对于以三次样条插值的等值线的数据压缩是行之有效的。 本文内容安排如下： 第一章：散乱数据插值与拟合；介绍了一些在地质统计学上常用的算乱数据插值方法，包括移动平均法、克里格法等。鉴于本文的需要，最后重点介绍了三次样条插值。 第二章：曲线的数据压缩；对已有压缩方法做了分类，并结合图示介绍了一些前人给出的多种压缩算法。 第三章：等值线的数据压缩；首先介绍了一种适用于三次样条曲线的数据压缩算法，其次给出了一种分段DP压缩算法，最后将二者结合实现了对等值线的数据压缩，并给出部分试验实例。