向量式资料结构.doc

向量式資料結構 一、資料模型 (data model) 資料模型(data model)：網格式、向量式、物件導向式(object oriented，另詳「代數資料模型」) There are different kinds of data models, such as: raster, vector, object oriented, and many-sorted algebraic data models. 向量式資料：以點、線、面、體的方式，於各圖層描述物件的圖形資料。 Vector data model uses points, lines, polygons, and volumes to describe geo-objects in terms of layers. 資料結構：資料的有序組成，俾能快速的資料新增、刪除、變更、查詢、計算。 Data structure is a well-organized way to describe objects, such that any object can be efficiently added, deleted, modified, queried, counted, and computed. 網格式資料(庫)也有多種資料結構，例如：二維矩陣、四分樹、階層式結構。 2-dimensional matrix, quad-tree, and hierarchy are some data structures of raster data. 吾人可以用不同的資料結構來表達(同一或不同的)TOPOLOGY(位相、拓樸)關係。不同的資料結構可能各有利弊，必須視不同的使用目的或分析目的，才能決定使用那一種資料結構。 We may use different data structures to describe topology. Each data structure has its own advantages and disadvantages. It depends on purpose to choose an appropriate data structure. 不同的GIS套裝軟體，視同潛在客戶的性質，使用不同的資料結構。是故，不同的GIS套裝軟體適用不同應用場合。內部的資料結構是商業機密。一般商業軟體僅提供外部資料格式，以供交換使用。 Depending on the need of most potential users, each GIS software choose a fixed internal data structure. Different GIS software packages are suitable for different users and applications. Internal data structure is a commercial secret. However, to exchange data among different software packages, external data exchange formats are defined. 二、資料變數的類型 (types of variables) 表面上看來，都是文數字。但是吾人必須仔細區分它究竟是那一種變數，或者吾人將之定義為何種變數。不同的變數類型，其變數值僅能容許不同的(順序、四則、…)運算。這與物件導向式的設計觀念相符。 We have to distinguish types of alpha-numerical variables. Different types of variables are allowed to operate with particular functions. (Object oriented programming takes the same approach.) As a matter of fact, the program designer decides which variable type will be used. nominal variables (名目變數)：以各種名稱描述物件。各名稱(變數值)之間沒有順序。可做集合之交集、聯集、差集等運算。 Nominal variable uses names to describe objects. There is no order among names. Set operations, such as intersection, union, difference, are legitimate for