Comparison of gis-vector and grid data structure

2.5 Comparison of vector and raster data structures

Early in the development of computer-aided cartography and GIS, vector processing technology was originally quoted, and raster data processing began in the middle of the 70 's. A few years ago, these two data structures were hard to be compatible, so they caused a lot of inconvenience to the use of data. In recent years, it has become more and more clear that the differences between grid and vector data structures have been regarded as important conceptual differences, which are in fact technical problems. The development of computer technology makes computing speed, storage capacity, spatial resolution of geographic data greatly improved. In order to utilize GIS more effectively, one of the problems people face is the choice of grid and vector data structure.

Comparison of 2.5.1 vector and grid data structure

The data range of geographic information system is very wide, the data is saved in various ways and the structure type is complex. The vector structure and grid structure of spatial data are two distinct methods of simulating geographic information system, which are different, complementary and mutually promoted. The detailed comparison between vector data and raster data structures is shown in table 2-5.

Table 2-5 Comparison of vector and raster data

Vector data	Raster data
Small Data storage	Large data storage capacity
High accuracy of space position	Low spatial position accuracy
The topological relationship can be fully described by the network connection method	Difficult to establish network connection relationships
Simple and easy output, precise and beautiful drawing	Fast output, but rough and unsightly drawing
Can retrieve, update, and synthesize graphs and their properties	facilitates surface data processing
Complex data structures	Simple data structure
Slow data acquisition	Fast access to large amounts of data
Difficulty in mathematical modelling	Easy Math Simulation
Multiple Map Overlay Analysis difficulties	Multiple map overlay Analysis is convenient
Cannot process digital image information directly	Ability to process digital image information directly
Spatial analysis is not easy to achieve	Spatial analysis is easy to do
Complex and vague boundaries are difficult to describe	Easy to describe complex, blurred boundaries
Higher cost of data output	Low technology development costs

2.5.2 vector, grid data integration

The new generation of integrated Geographic information system requires the unified management of graphical data, attribute data, image data and digital elevation model (DEM) data, called Siku unity. The unified management of graphic data and attribute data has made breakthrough progress in recent years, and many GIS software vendors have launched their own spatial Database engine (SDE) to solve the integrated management of graphical data and attribute data. and vector and raster data, according to the traditional concept, is considered to be two kinds of completely different data structures, when using them to express space objects, for linear entities, people are accustomed to using vector data structure. For polygon entities, in the vector-based GIS, the boundary expression method is used, but in the raster-based GIS, the meta-subspace is generally used to fill the expression. Therefore, it is believed that the linear entity represented by the vector method can also be represented by the meta-subspace filling method, that is, when digitizing a linear entity, the resulting raster is recorded in addition to the original sampling point. Similarly, each polygon figure records the raster contained in the middle, in addition to its polygon boundaries. On the one hand, it retains all the properties of the vector, gathers all the position information directly to the target, and establishes the topological relation, on the other hand, it establishes the relation between the grid and the figure, that is, any point on the path is directly connected with the target. In this way, both the vector and raster properties are maintained, and the vectors are unified with the raster, which is the basic concept of the integrated vector and raster data structure (Shihung, 2003).

2.5.3 vector data and grid data selection

According to the above comparison, in the process of GIS establishment, the appropriate data structure should be selected in the vector and raster data structures according to the application purpose, actual application characteristics, possible data precision and GIS software and hardware configuration. Vector data structure is one of the most familiar forms of graphic expression, and for line map, it is often compared with raster data to save storage space. Interconnected line networks or polygon networks can only be achieved with vector data structure patterns, so the vector structure is more conducive to network analysis (transportation networks, supply, drainage networks, gas pipelines, cables, etc.) and mapping applications. The data represented by vector data is of high precision, and it is easy to attach a classified description of the properties of the cartographic object. Vector data can only be exported on a vector data plotter. At present, analytic geometry is frequently used in the processing of vector data, for some direct and point-related processing and there is a ready-made mathematical formula for the operation of individual symbols, using vector data has its unique convenience. Vector data facilitates the creation of individual cartographic objects and facilitates the storage of relationship information between graphical elements.

Grid data structure is a kind of image data structure, which is suitable for remote sensing image processing. It has a simple, intuitive and strict correspondence with the spatial distribution characteristics of cartographic objects, which is of great potential for the spatial location of cartographic objects, and provides the possibility for application of machine vision, which is the most convenient for detecting the position relation between objects. The raster selection scheme is often used in the calculation of polygon data structures, and in many cases, the grid scheme is more efficient. For example, polygon circumference, area, sum, calculation of averages, radii from a point of view, etc. are all reduced to simple counting operations in the raster data structure. And because grid coordinates are rules, deleting and extracting data can be done by location-determined windows, which is much more convenient than vector data structures. Recently, the raster algorithm based on vector data structure shows that there is a more effective method than previously thought to solve the problem of some raster structures. For example, an issue where the data storage of a raster structure is too large can be reduced by the compression method presented in section 2.1 of this chapter.

Both the grid structure and the vector structure have some limitations. Generally speaking, a large-scale small proportion of natural resources, environment, agriculture, forestry, geology and other regional issues of research, urban planning phase of the strategic layout of the study, the use of raster model is more appropriate. Urban zoning or detailed planning, land management, utility management and other aspects of the application, vector model is more appropriate. Of course, you can also mix the two models and display two different ways of mapping on the same screen.

At present, the developers and users of GIS are actively studying the mutual conversion technology of these two kinds of data structures, and have developed the software to convert the raster data structure and the vector data structure to each other. Vector-to-raster conversion is simple, and there are many well-known programs that can accomplish this conversion. And there are many display screens that can automatically complete the conversion work. Raster-to-vector conversions are also easy to understand, but the exact algorithm is much more complex. To realize the mutual transformation of two data structures can greatly improve the generality of GIS software, in recent years, some people in the experiment with a software to implement both raster and vector two models to facilitate user use.

Ref

2.5comparison of vector and grid data structures

Comparison of gis-vector and grid data structure