Online Distributed GIS System

Based on Web Service and ArcGIS Server

Jian Huang

Abstract:

The trend of the whole IT industry is moving to distributed system. This project focuses on establishing an online distributed GIS system which can process popular GIS data format using web based GIS applications. The project attempts the utilize web service to implement advanced online GIS tools and demonstrate the interactive system which allows user upload data or develop analysis method.

Key words: ArcGIS Server, web service, ArcObjects, OGC, distributed GIS system

Introduction

Traditional GIS is closed, centralized system. Users cannot exchange their data or method in real time manner. Distributed GIS services focus on the online processes of information services and task-oriented internet GIS applications.

Because distributed GIS is a distributed system, databases and application programs link with each other dynamically. This dynamic nature allows distributed GIS to be more capable in linking with real time information, such as real time satellite images, traffic movements and emergency response information.

In short, distributed GIS is a special type of GIS tool that uses internet as the major means to access and transmit distributed data and analysis tools. It’s object oriented, distributed and interoperable. The end user does not necessarily need to have GIS data and software installed in his or her local computer, because all data and analysis modules can be available on network servers. Users can request and get response according to their demand. At the same time, if the user want to use their own data or analysis module. He can upload his data or develops his own analysis method.

It solves the different data format problem, it solves different operation system or GIS software conflict problem.

Traditional GIS system has difficulty in providing flexible, friendly GIS solutions for users. Different GIS systems have unique functions and data formats, which can not be shared. By making advantage of distributed GIS system, the huge amount of data can be managed and upgraded by server side. Meanwhile it makes the development of GIS system into a sustainable strategy. Data sets and GIS operators will be dynamic, movable, and interoperable on the internet or intranet.

The advantages of online distributed GIS system are:

•It is a Browser and Server (B/S) system. The user only needs a low end computer with the internet access. It saves the cost ofGIS workstation hardware and software.

•When it has problem, all the upgrade and maintenance can be done in the server side. The support analyst doesn’t need to be in the client side. It saves the cost for maintenance.

•No license number limitation.

•The developer can design user interface for specific purpose.

•Plenty of data and analysis methods exchange.

The disadvantages are:

•It requires intensive programming and involves many IT technologies: XML, .NET, SOAP, ArcObjects, etc.

•It requires high speed internet.

The two major platform and technologies that the project adopts are ArcGIS Server and web service. The following introduce them in a short manner.

ArcGIS Server is a platform for building enterprise GIS applications that are centrally managed, support multiple users, include advanced GIS functionality (versus ArcIMS).Developers can use ArcGIS Server to build Web applications, Web services, and other enterprise applications.ArcGIS Server could be used to build desktop applications as well that interact with the server in client/server mode.

A web service is a software system designed to support interoperable machine-to-machine interaction over a network. It has an interface described in a machine-processable format (specifically WSDL). Other systems interact with the Web service in a manner prescribed by its description using SOAP messages, typically conveyed using HTTP with an XML serialization in conjunction with other Web-related standards.(

Shortly, software applications written in various programming languages and running on various platforms can use web services to exchange data over computer network like the internet (http// The web service connects developers on the internet. Developers can share data and methods through protocols. It improves distributed access to software capabilities and removes many of the problems caused by incompatible platforms. You can create new web services by combining different web services to provide an integrated service.

Literature Review

Putz and Huse create the early popular internet gis services using the web browser via HTML format.Plewe uses the term distributed geographic information to refer to the use of internet technologies to distribute geographic information in a variety of forms, including maps, images, data sets, analysis operations, and reports.Dixon states the necessary technologies for online GIS implementation. Huang and Lin create a virtual reality modeling.Huang and Worboys create hydrology model in web based GIS tools.

Slater analyzes transportation in Queensland.UTD GIS group implements a Waste-to-Energy Strategic Geographic Planning Tool. Other originations put efforts in trying to establish standard GIS web service, such as Open Geospatial Consortium (OGC)
The OGC is a non-profit, international, voluntary consensus standards organization that is leading the development of standards for geospatial and location based services. Through its member-driven consensus programs, OGC works with government, private industry, and academia to create open and extensible software application programming interfaces for geographic information systems (GIS) and other mainstream technologies. Asrar and Dozier introduce Earth Observation System (EOS).Evans introduces OGC Web Coverage Service (WCS).Yang and Whitesidedetails all the parameters for OGC Web Image Classification Service (WICS). Di, Yang, Deng, and McDonald create an interoperable, On-demand Geospatial Data Access and Services Based on OGC Web Coverage Service (OWS) Specification. ESRI provides ArcWeb Service to make distributed GIS system easier.

CaseImplementationI. Extending User Interface:

ArcGIS Server provides some template to meet the simple GIS analysis. This case is shown as below. The scenario is if you have several options for your apartment, the surrounding environment is the key consideration for the purchase or rent. You can select a target and set the buffer radius, such as 2 miles, and set the selected features as Parks, se the method as intersect, then click “Find”, the result is shown as figure 2.

Fig. 1 User interface of the buffer and select template

Fig. 2 Buffer and select result

Case Implementation II. Extending functionalities:

The purpose of this case is to assign a complex digitizing job to many technicians. The old way is to let them go to the office with the digitizing software like ArcView and combine all digitized data into a entire shapefile or geodatabase. That raises some problems such as management, data format conversion. Now they can work at their home and collaborate at the same time since there is a common geodatabase behind the application. (Some code is from ESRI).

Fig.3 User interface of digitizing application

  1. Zoom in to the area you want to digitize. Select the building type from the drop down list, such as SportsField.
  2. Click the Start a Edit Sessionbutton, it makes the geodatabase ready for digitizing. Then click the polygon button to start the digitizing, double click to finish the polygon. You can see a polygon with the label appears.
  3. If you are satisfied with the result, then click Stop editing and save your edits button to save the digitizing into the geodatabase.
  4. Click the Download button under the table of content and save the geodatabase into your local system for the future analysis or presentation.

Case Implementation III. Data customization:

One of the big disadvantage of the online GIS application is that the server provides the data and users have no privilege to use their own data. For example, when you are using Yahoo or Google map to search the location by address, all the data is from Yahoo or Google’s server. But you cannot expect they upgrade the data in real time, so the best way to solve this problem is to allow user to upload data.

There is a road map of city of Richardson, you can find the location by address. The ArcGIS Server contains the basic geocoding template.

Fig.4 Geocoding application with data customization

  1. Input the address and zip code (optional), such as 2200 waterview pkwy, 75080, click Location button, one result appears in the table, click Map button, the location will display in the map.
  2. So far, the function of address finder is similar to Yahoo or Google map. If the users think the road map is not accurate or they want to find location in other cities, then the data customization is very important.
  3. Click the Remove button to remove the current road map.
  4. Download the Plano road map from This zip file contains a road map shapefile of city of Plano.
  5. Click Browse button, navigate to the zip file that you just download, and click upload. The new map will show as fig.5.

Fig.5 Data customization

  1. Input the address in city of Plano, such as 1500 Coit rd, 75075, the new location will display in the map view.

So far, the online distributed GIS system has user input option, download option. The server acts as data processor. The online distributed GIS system works as convenient as local desktop system.

Case Implementation IV. Web Service:

As we discuss before, web service make the developing software much easier, this online unsupervised classification case shows the power of web service.

The old way of digital image processing is to buy digital image software. The problem is: 1. most users only need some specific functions. But you have to pay big money for the whole software package. 2. If too many users in the same organization want to use the software at the same time, it has license number limitation problem.

Anunsupervised classification web service which was developed by Geobrain team in the GeorgeMasonUniversity is consumed. It can classify remote sensing image withEOS-HDF format. As a developer, I don’t need to write the classification algorithm from scratches.

The Earth Observing System (EOS) is the centerpiece of NASA's Earth Science Enterprise (ESE). It is composed of a series of satellites, a science component, and a data system supporting a coordinated series of polar-orbiting and low inclination satellites for long-term global observations of the land surface, biosphere, solid Earth, atmosphere, and oceans. EOS will enable an improved understanding of the Earth as an integrated system.

The classification is based on ISODATA method. ISODATA represents a comprehensive set of heuristic procedures that have been incorporated into an iterative classification algorithm (Jensen, 2005).

  1. Download the sample data from Open the image by ERDAS IMAGINE or ENVI.

Fig.6 Sample EOS-HDF remote sensing image

  1. Navigate to the downloaded file by clicking the Browse button.
  2. Specify the Number of Clusters as 3, select the Type of Output as PNG or HDF-EOS. PNG will return the result as a picture, while HDF-EOS will return a .hdf file as the result which can be opened using digital image software like ERDAS IMAGINE.
  3. Click Classify button. The result shows as fig.7.

Fig.7 Unsupervised classification result using web service

Only hundreds of lines code implements the online unsupervised classification. Web service increases the developing efficiency dramatically.

Case Implementation V. Interoperable Web Services Chain:

We can create sophisticated application by using web services chain, which means combine many web services together.

We introduced unsupervised classification, then it raises another question naturally, how to get data efficiently. The scenario utilizes ESRI ArcWeb Service and Open Geospatial Consortium web services to search data.

Fig.8 Address finder and map image

  1. Input the address or intersection, set the X Range and Y Range as 0.2, which means the result map area will be 0.4*0.4 surround the location.
  2. Click Get Map Area and Image button, then the map will show as below. The red star is the target location. And the corresponding log/lat of the address location and map area appears.
  3. You can zoom in or zoom out to get more detailed or rough information.
  4. You can choose other data source, right now, only few of them are available since others are commercial service.

Fig.9 Returned image

  1. Click Get Landsat Data button. All Landsat data with the area intersecting with the map area will be searched and returned. The description column describes the basic information about the data, including data type, path, row, band and date. Click image/tiff link, the remote sensing data can be saved to your local system. GMU developer team creates the data search service.

Case Implementation VI. A Suitability Model:

This case introduces a real world project which solves a capability problem. Capability is referred as physical potential of land unit to support natural phenomenon. For example, an agriculture engineer is interested in the good place where is suitable for crop growth. The factors should include humidity, temperature, radiation and precipitation.

Fig. 10 Suitability model based on raster analysis tools

  1. The server provides humidity, radiation and temperature data, but it’s not enough for the analysis. Download the precipitation data from which is a raster file represents the precipitation data. Certainly you can upload any data that you think correct for the model. The only requirement is that it must be a zip file, and its name must be as the same as the raster file.
  2. Click Browse button and select the file you just downloaded, and click Upload button. The precipitation data will display in the map view.
  3. Click the button with letter Rwhich indicates reclassify. Then a window pops up. Select the target layer, such as humidity. Click the plus symbol to add a row of reclassification schema. Click OK, the result appears in the map view with the name Re_humidity.

Fig.11 Reclassification schema

  1. The new generated raster file is shown as below, the cells originally with the value from 638 to 1200 are assigned the new value 1. The same logic, other cells change value according to the schema. The new data represents the index of effect for the crop growth. For example, the area with value 1 is not suitable compared to those areas with higher value.

Fig.12 Reclassified result

  1. Apply the same step for other layers, such as radiation layer. Be cautious that the from value of the first row should smaller than the Low value of the original raster layer, and the to value of the last row should be higher than the high value.
  2. Now you have Re_precipitation, Re_humidity, Re_radiation and Re_temperature. The capability model should consider the multiple effects of all the layers. Click Re_humidity and click Add as a operator button . Then the layer name appears in the operation expression textbox.
  3. Click plus symbol, then click another factor layer, such as Re_radiation and add it to the operation expression box.
  4. Click equal symbol , after a while, the result will display. You can try any combination with weights, such as [Re_precipitation] * 0.3 + [Re_humidity] * 0.7, or [Re_precipitation] *0.2 + [Re_humidity] * 0.3 + [Re_radiation] * 0.4 + [Re_temperature] * 0.1.
  5. If you are not satisfied with the result, select it in the table of content and click Remove button and start a new combination.
  6. The result could be saved into your local computer. Select the data format, such as TIFF, and click Savebutton. A zip file with a long GUID name is ready for downloading. You might need it for further analysis or making a map.
  7. You can select any layer and click Move up or Move down button to change the order of layers in the table of content.
  8. Select any raster layer, and click Statistics button, then the basic statistical parameters will be calculated.

By the way, the reclassification and raster calculator tools are for general purpose. The capability model is only one possible application using those tools; you can apply them to any raster analysis.

Conclusion

This project creates an online GIS system based on advanced IT technologies, which can process vector, raster and remote sensing data analysis. The server can provide analysis methods and data. It also allows users to use their own data. The project adopts web services to implement advanced GIS functions and data search tool.

Reference:

•Andrienko, G., Andrienko, N., Voss, H., Carter, J. (1999). Internet mapping for dissemination of statistical information. Computers, Environment and Urban Systems, 23: 425-441

•Asrar, G., and Dozier, J., 1994. EOS: Science Strategy for the Earth Observing System. American Institute of Physics Press, Woodbury, New York11797.

•Di, L., W. Yang, D. Deng, and Ken. McDonald, 2002. “Interoperable, Personalized, On-demand Geospatial Data Access and Services Based on OGC Web Coverage Service (OWS) Specification”, Proceeding of NASA Earth Science Technology Conference, CDROM, Pasadena, California. 3pp.