What Is a Spatial Database?

Spatial database refers to the sum of application-related geospatial data stored on a computer physical storage medium by a geographic information system, and is generally organized on the storage medium in the form of a series of files with a specific structure. The research of spatial database began in the field of map drawing and remote sensing image processing in the 1970s, and its purpose was to quickly use satellite remote sensing resources to quickly draw various economic thematic maps. Due to the many shortcomings of the traditional relational database in the representation, storage, management and retrieval of spatial data, the database research field of spatial database has been formed. The traditional database system only targets simple objects, and cannot effectively support complex objects (such as graphics and images). A spatial database is a collection of data about the characteristics of certain spatial features in a certain area. It is the sum of application-related spatial data stored on the physical medium by GIS. ^[1]

Spatial database

The spatial database is

1,

Databases have various organizational forms due to different application requirements. The design of the database is to determine the optimal data model, processing mode, storage structure, and access method in a given application environment according to different application purposes and user requirements, and establish information between geographical entities that can reflect the real world. The database can meet the user's requirements and be accepted by a certain DBMS. At the same time, it can achieve the system goals and effectively access and manage the data in the database. In short, database design is the process of abstracting application data that exists in a certain range in the real world into a specific structure of a database.

The design of the spatial database refers to the entire process of establishing a spatial database on the basis of a current database management system. It mainly includes three parts: requirements analysis, structure design, and data layer design.

1. Demand analysis

Requirements analysis is the basis for the design and establishment of the entire spatial database. The main tasks are as follows:

1) Survey user needs :

Understand the characteristics and requirements of users, and obtain a consistent view of the needs of designers and users.

2) Collection and analysis of demand data :

Including information requirements (information content, characteristics, data to be stored), information processing requirements (such as response time), integrity and security requirements, etc.

3) Preparation of user requirements statement :

Including the objectives, tasks, specific requirements specifications, system functions and performance, and operating environment of the requirements analysis are the final results of the requirements analysis.

Requirement analysis is a very technical task and should be completed by experienced professional and technical personnel. At the same time, the active participation of users is also very important.

Complete the selection of data sources and evaluation of various data sets during the requirements analysis phase

2.Structural design

Refers to the spatial data structure design. The result is a reasonable spatial data model, which is the key to the spatial database design. The more the spatial data model can reflect the real world, the more the application system generated on this basis can better meet the user's requirements for data processing.

The essence of spatial database design is the process of expressing geospatial entities in a database system in a certain organization form, that is, the modeling of spatial entities in geographic information systems.

1) Conceptual design

Conceptual design is the understanding and abstraction of the intricate real world, and finally forms the model required for the spatial database system and its application system.

Specifically, it analyzes and organizes the information and data collected during the requirements analysis phase, determines the geographic entities, attributes, and the relationships between them, combines the partial views of each user into a total global view, and forms a computer-independent reflecting user. Perspective conceptual pattern. The concept mode has nothing to do with the specific DBMS, has a stable structure, and can better reflect the user's information needs.

The most powerful tool for representing conceptual models is the ER model, which is the entity-connection model, which includes three basic components: entity, connection, and attribute. Use it to describe the real geographic world without having to consider computer-related issues such as information storage structure, access path, and access efficiency. It is closer to the real geographic world than ordinary data models, and is intuitive, natural, and rich in semantics. Features have been widely used in geodatabase design.

2) Logical design

Based on the concept design, the process of converting a conceptual model into a data model supported by a specific DBMS according to different conversion rules, that is, exporting the logical structure (or external model) of the geodatabase that a specific DBMS can handle, including determining data items, records And records, security, integrity, and consistency constraints. Whether the derived logical structure is consistent with the conceptual model and whether it can meet user requirements, its function and performance must be evaluated and optimized.

The main process of transforming from E-R model to relational model is:

determine the main keywords of each entity;

Determine and write the data relationship expression between the internal attributes of the entity, that is, a certain data item determines another data item;

Use the entities in the data relationship expression after redundancy processing as the corresponding primary keywords

Form a new relationship according to , .

After completing the conversion, analyze, evaluate and optimize.

3) physical design

Physical design refers to the effective implementation of the logical structure of the spatial database on physical storage, and the physical storage structure of the data on the medium is determined. The result is a storage mode (internal mode) for the geodatabase. The main contents include determining the record storage format, selecting the file storage structure, determining the access path, and allocating storage space.

The physical design will have a great impact on the performance of the geodatabase. A good physical storage structure must meet two conditions: first, the geographic data occupies a small storage space; second, the database operation has the highest possible processing speed. After the physical design is completed, performance analysis and testing are performed.

The physical representation of data is divided into two categories: numeric data and character data. Numerical data can be expressed in decimal or binary form. Binary forms usually take up less storage space. Character data can be expressed in the form of character strings, and sometimes the storage of code values can be used instead of the storage of character strings. In order to save storage space, data compression technology is often used.

The physical design is largely related to the database management system chosen. In the design, the functions provided by the system should be selected according to the needs.

4) Data layer design

Most GIS organizes data into different data layers according to logical types. The data layer is an important concept in GIS. GIS data can be divided into various logical data layers or professional data layers according to the logical relationship or professional attributes of spatial data, which is similar to the superposition of pictures in principle. For example, topographic map data can be stored separately in various layers such as landforms, water systems, roads, vegetation, control points, and residential areas. The data of the topographic map is synthesized by superimposing the layers. When performing spatial analysis, data processing, and graphic display, only a few corresponding layers of data are often required.

The design of the data layer is generally based on the professional content and type of data. The type of professional content of the data is usually the main basis for data layering, and the relationship between the data must also be considered. If it is necessary to consider the shared boundary of two types of objects (the overlap of road and administrative boundaries, the overlap of rivers and plot boundaries), etc., the relationship between these data should be reflected in the data hierarchical design.

Because different types of data have the same application functions, they are often used at the same time in analysis and application. Therefore, such requirements should be reflected in the design, and these data can be used as a layer. For example, polygonal lakes, reservoirs, linear rivers, ditches, point-shaped wells, springs, etc. are often used simultaneously in the application of GIS, so they can be used as a data layer.

5) Data dictionary design

The data dictionary is used to describe the overall structure, data content, and definition of the database. The contents of the data dictionary include: 1) The overall organizational structure of the database and the framework of the overall design of the database. 2) Definition of detailed content and structure of each data layer, definition of data naming. 3) Metadata (data about data is a description of the content, quality conditions and operation process of a data set).