الفهرس Only 14 pages are availabe for public view
 |  | from | 126 |  |  |
| | | from | 126 | | |
Abstract
The development of Geographic Information Systems (GIS) has been promoted by the development of Information Technology (IT) networking, and software engineering.
Due to the popular use of the Internet and the progress of telecommunications, GIS is shifting into a new direction, Distributed GIS (DGIS). In distributed GIS environments, users may need to interact with heterogeneous data models and different command syntax in different GIS software that deal with geo-information as one of the most complicated information types stored in computer systems. In this Thesis the mobile agent is used to reduce the user work overhead and information overloads. In this context, intelligent mobile agents provide services in searching, filtering geo-data and geo-processing services, converting GIS data from one format to another. The proposed system has several potential advantages: reduced network traffic, decentralization, increased robustness and fault-tolerance.
The proposed system is composed of 4 tiers that support advanced search, filtering, queering functions, and interpreting GIS data for distributed GIServices. The .Net framework, mobile agent and GIS web Services are used to implement this system that involves; client sites, sever sites and mobile agents roaming on the internet/intranet for retrieving information on behalf of the users.
In this research, implementation of the proposed system is applied on two cases studies, which were taken upon realistic problem facing our country in the mean while.
Spodoptera littoralis prediction and management model” is developed as a first experimental case study. The model calculates the minimum and the maximum air temperatures that are derived from satellite images. Based on these values, degree-days (DD’s) required for the completion of pest stages are calculated. Acquiring the pest stage from the user and calculated the affected zone and predict the date of inspection. Comparing the result with an experimental result 95%.
By expanding the implementation of this model to other pest agricultural problems, this could help in reducing the food crises facing our country for the next years.
Monitoring the Urban Growth on Agricultural Land Model; considering Ismailia governorate as an example; is the second experimental case study. First the model classifies the satellite images for each of the start and end dates into 5 classes (Barren land, Urban, Agriculture, Water and Wetland). As a result of comparing the images, a change image with positive and negative changes is displayed along with a change matrix that describes the total change for each class and its percentage. The change percentages in the classes are -0.41%, 0.23%, -1.59%, -0.23%, 0.20% respectively. With applying this model to other governorates will help decision maker to define the actual size of urbanization problem over the agricultural land.
Keywords: GIServices, GIS Meta Data, Mobile Agent, Information Technology.