Providing a Ubiquitous Data Model in Order to Extract Spatial Relations

Today, ubiquitous systems has been growing to become the new generation of Geospatial Information Systems (GIS). In Ubiquitous GIS, servicing capabilities to any user, at any time, in any location, using any device, and in any condition are provided. Advances in Information Technology (IT) industry have led to the advent of low-cost three dimensional data acquisition technologies (e.g. Microsoft Kinect). However, there are some deficiencies and limitations in utilizing these novel technologies in Ubiquitous GIS environments. Among them is the lack of information about the three dimensional topological relations between identified objects. Moreover, the current data models for the extraction of topological relations in ubiquitous environments are not sufficient. Therefore, the aim of this study is to extract the topological relations between objects sensed by Microsoft Kinect by designing a novel Ubiquitous GIS Data Model (UGDM). To obtain the extent of each object, the bounding box algorithm has been employed. The results of analyses on the acquired data from Kinect sensor show the success of the proposed data model in topological information .Ability to use spatial information at any time and places based on the use of IT and different infrastructures like sensor network (SN), Internet, and communication devices used in the system must be arranged by a ubiquitous data model to make different services at any time, any place, by any device and data for any user. Spatial relationships between objects, containing the fundamental information relating to the environment. Among these spatial relationships can be noted such as directional relationships, like: “left”,”right”,” above”, and “bottom” or distance based relations such as: “near” or “far”. The topology relationships used more in GIS desired location can be also noted such as: “inside”, “intersection”, “touch”, and “outside” and etc. Extraction of relationships between different features are done on the basis of data modeling techniques. Since the traditional data models don’t cover capabilities of the ubiquitous computing tools used in ubiquitous GIS to extract 3D relationships, we discuss the creation of facilities for the extraction of spatial relationships in ubiquitous GIS. A new generation of GIS is ubiquitous GIS, which in this generation service functionality to any user, at any given time and place. With the advancement of new technologies in capturing 3D point clouds of the environment, general changes are created in ubiquitous computing. The sensors used to obtain environmental information is provided in this generation are often inexpensive sensors. One of the important issues associated with this generation is the extraction of 3D topological relationships to interact with the environment better, which these sensors don’t have this ability to extract 3D topological relationships. To provide such a ubiquitous data model it is required to use software and hardware standards in order to exchange information in different parts of the ubiquitous model properly. The goal of this article is providing a ubiquitous data model based on the predefined standards to extract spatial relations. Spatial relations include directional, projection, distance and topological manners examined in this article. Because there is no unique method for extracting different spatial relations topological relations. Developing a spatial language as Ubi-OCL for defining different elements used in the data model provides an interoperable connection between the user and the data model. In this language the user can select the type of sensor and spatial relations to be extracted. For evaluating the presented data model, topological relations extracted will be presented in the article. As conclusion this article shows the different results with different sensors, the bounding box method is used for processing Kinect’s point cloud for extracting spatial relations according to provided data model.