Current status of geographical information system integration

GIS is a special information system that processes geographical information. The difference between geographical information and other information is that its data contains both attribute data and spatial data, both of which are equally important. Geographic information systems are generally defined as information systems used to collect, store, manage, analyze and express spatial data. From this perspective, GIS includes the entire process of spatial data acquisition, processing and application. It is an integrated system from the beginning.

GIS integration refers to the integration of GIS itself in a narrow sense, and in a broad sense includes the comprehensive integration of data collection systems, processing analysis systems, and application systems related to geospatial information. In practical applications, it may also include the integration of office automation systems, communication systems, command and dispatch systems and other related systems according to user needs.

The integration of GIS itself includes data integration, model integration and software integration. Data integration refers to how geographical information data from different regions, different times, and different contents (attributes) form a unified whole and are appropriately segmented according to certain rules to form a physically distributed, logically centralized, practical and Efficient distributed spatial database system. Model integration refers to how to establish models that reflect spatial regularities, and organically combine these models with database platforms and supporting functional software platforms to realize the management of model libraries, method libraries, and knowledge bases, and integrate existing models and methods into and build new models based on knowledge. Software integration mainly refers to the integration of GIS functional software support platforms, including integration between modules within the same platform and integration between functional modules of different GIS platforms, and may also include integration with other types of software. There are still some problems in integration technology. With the advancement of computer hardware and the development of software technology, many of these problems will be gradually solved. However, the core problem of GIS integration still exists, which is the establishment of a unified spatiotemporal data model. , based on this data model, unified logical and physical segmentation of space-time bodies and how to establish a problem-oriented spatial analysis model.

In recent years, some scholars and experts have proposed "3S integration" or "5S integration". "3S integration" refers to the integration of GIS, RS, and GPS, and "5S integration" also includes digital photogrammetry system (DPS) and expert system (ES). RS is closely related to GIS. RS has always been regarded as an important source for GIS to quickly obtain information, while GIS is considered to be a deepening of the utilization of RS information. Some people have proposed the integration of RS and GIS a long time ago, and proposed separate but parallel methods. There are several modes such as combination, seamless combination and overall combination, but there has been little progress in practical applications. The general mode of "3S integration" is that RS collects area information, GPS collects real-time point data, provides data sources for GIS, and uses GIS for spatial data management and spatial analysis. The introduction of DSS and ES aims to deepen the application of spatial data. DSS mainly deals with semi-structured and unstructured problems. Its purpose is to assist decision-making rather than replace human functions. ES simulates the analysis and decision-making process of experts, replacing the role of humans. The acquisition, analysis and application of spatial information is a connected, cyclical process. The plans and decisions generated by the application of spatial information act on spatial objects to change their states, starting a new cycle of spatial information acquisition, analysis and application. In this model, "5S integration" forms an organic whole through feedback and coordination mechanisms, representing a new direction in geographic information system integration.

In practical applications, aerospace remote sensing information, surveying and mapping information and ground observation data are used as data sources, and GIS spatial analysis functions are used to establish professional evaluation and analysis models for spatial information management and decision-making. The integrated system has played an important role in the fields of dynamic monitoring of resources and environment, crop yield estimation, monitoring and assessment of major natural disasters, ecological network research, urban planning and management, etc. There are many successful systems at home and abroad.

There are methodological problems in current GIS system integration. A preliminary summary includes the following aspects that are worth noting:

(1) The theoretical foundation of information integration is weak. The types, sources, representations and applications of geographic information are becoming increasingly diverse. Although GIS integrated systems have made great progress in the practice of distributed flat network extension (WEBGIS), three-dimensional extension (3D GIS), time extension (TGIS), and attribute extension (multi-source data fusion), currently in GIS There is also a lack of effective expression of a geographically unified space-time body.

(2) The methodology of GIS integration is lacking. There is a lack of consistent understanding of the concept of geographic information integration, and there is a lack of in-depth theoretical research on how GIS is integrated, what is included in GIS integration, and the architecture of the integrated system.

(3) There is more mechanical integration and less organic integration. Many integrated systems are just a simple stacking of subsystems, resulting in redundant and inconsistent data and functions, making them unable to be put into business operation. This is mainly due to the lack of comprehensive understanding of system integration. Generally speaking, system integration is not just functional integration, but also includes resource integration, information integration and personnel integration. The integrated system should not only have the functions of each subsystem, but also provide system feedback and collaborative functions, while providing the possibility to add new functions in the future and form an organic whole.

(4) The integrated architecture is unreasonable. GIS integration is generally the unification of multiple hardware platforms, multiple operational collaborations, multiple network platforms, multiple support software, and multiple application integrations. One or several commercial software alone cannot meet the needs of the overall system, and it is based on commercial Centralized systems supporting software have many problems in system expansion, upgrades and maintenance, and all applications cannot be separated from the system. From a centralized system to a distributed system based on C/S structure, commercial software is regarded as a server that provides specific functions. Various commercial software modules and self-developed modules can be combined as needed, and can be easily upgraded and replace.

(5) Poor flexibility. GIS applications are generally macro-dynamic and complex systems. However, users of current integrated systems can generally only complete prefabricated functions according to the menus provided by the system, and cannot quickly establish application-oriented analysis models according to actual needs. This is mainly due to the fact that the integrated system does not have a unified information view, lacks a database integration platform and a model library integration platform, and lacks convenient, flexible and visual modeling tools.

(6) Lack of standards. Most of the early GIS integrated systems were research and experimental systems, and there was no unified standard. But standardization is the basic guarantee for the success of large and complex systems. As more and more spatial data standards are formulated, drawing on and integrating CIMS standards, software and hardware standards to formulate GIS integration standards is an important part of the GIS integration system's transition from experimental research to operational operation.