Marco Moreno

Ontology-Driven description of spatial data for their semantic processing

We use ontologies in this paper to search for alternative representations of geographic objects thus providing a description of these objects in cartographic vector maps. We define ontologies based on two types of concepts (“terminal” and “non-terminal”) and two kinds of relations (“has” and “is-a”). These are the basic elements used to describe a map. We also present a case study in which an ontology for topographic maps is created. Our approach is oriented towards solving heterogeneity and interoperability issues in GIS.

Geomorphometric Analysis of Raster Image Data to detect Terrain Ruggedness and Drainage Density

We present an approach to identify some geomorphometrical characteristics of raster geo-images. The identification involves the generation of raster layers, topographic ruggedness and drainage density. The topographic ruggedness is used to express the amount of elevation difference between adjacent cells of Digital Elevation Model (DEM). The topographic ruggedness is presented by means of Terrain Ruggedness Index (TRI). The densities layers are obtained by Spline Interpolation Method. These layers are used to represent the amount of geographic linear objects. The algorithm has been implemented into Geographical Information System (GIS) – ArcInfo, and applied for a GIS of Tamaulipas State, Mexico.

Geospatial information integration based on the conceptualization of geographic domain

Geospatial information integration is not a trivial task. An integrated view must be able to describe various heterogeneous data sources and its interrelation to obtain shared conceptualizations. Up-to-date, there are different and public ontologies for many domains and applications. Ontology engineering is rapidly becoming a mature discipline, which has produced various tools and methodologies for building and managing ontologies. However, even with a clearly defined engineering methodology, building a large ontology remains a challenging, time-consuming and error-prone task, since it forces ontology builders to conceptualize their expert knowledge explicitly and to re-organize it in typical ontological categories such as concepts, properties and axioms. In this paper, an approach to conceptualize the geographic domain is described. As a result of this conceptualization, we propose a semantic method for geospatial information integration. This consists of providing semantic descriptions, which explicitly describe the properties and relations of geographic objects represented by concepts, while the behavior describes the objects semantics. Summing up, this work presents a methodology allowing integrate and share geospatial information. It provides feasible solutions towards these and other related issues such as compact data by alternative structures of knowledge representation and avoids the ambiguity of these terms, using a geographic domain conceptualization. The general vision of the paper is to establish the basis to implement semantic processing oriented to geospatial data. Future works are focused on designing intelligent geographic information systems (iGIS).

Landform Classification in Raster Geo-images

We present an approach to perform a landform classification of raster geo-images to obtain the semantics of DEMs. We consider the following raster layers: slope, profile curvature and plan curvature, which have been built to identify the intrinsic properties of the landscape. We use a multi-valued raster to integrate these layers. The attributes of the multi-valued raster are classified to identify the landform elements. The classification approach is used to find the terrain characteristics of the water movement. Moreover, we describe the mechanisms to compute the primary attributes of digital terrain model. The method has been implemented into Geographical Information System-ArcInfo, and applied for Tamaulipas State, Mexico.

Simultaneous Segmentation-Recognition-Vectorization of Meaningful Geographical Objects in Geo-Images

We present an approach to color image segmentation by applying it to recognition and vectorization of geo-images (satellite, cartographic). This is a simultaneous segmentation-recognition system when segmented geographical objects of interest (alphanumeric, punctual, linear, and area) are labeled by the system in same, but are different for each type of objects, gray-level values. We exchange the source image by a number of simplified images. These images are called composites. Every composite image is associated with certain image feature. Some of the composite images that contain the objects of interest are used in the following object detection-recognition by means of association to the segmented objects corresponding “names” from the user-defined subject domain. The specification of features and object names associated with perspective composite representations is regarded as a type of knowledge domain, which allows automatic or interactive system’s learning. The results of gray-level and color image segmentation-recognition and vectorization are shown.

Geospatial Information Integration Approach Based on Geographic Context Ontologies

Geospatial information integration is not a trivial task. An integrated view must be able to describe various heterogeneous data sources and its interrelation to obtain shared conceptualizations. In this work, an approach to geospatial information integration based on the conceptualization of the geographic domain is described. As a result of this conceptualization, we propose a semantic method for geospatial information integration. This consists of providing semantic descriptions, which explicitly describe the properties and relations of geographic objects represented by concepts, while the behavior depicts the objects, semantics. Also, this method allows us to compress and share geospatial information by means of alternative structures of knowledge representation. Thus, it avoids the ambiguity of the terms, using a geographic domain conceptualization. The general vision of the paper is to establish the basis to implement semantic processing oriented to geospatial data. Future work is focused on designing intelligent geographic information systems (iGIS).

Semantic Supervised Clustering Approach to Classify Land Cover in Remotely Sensed Images

GIS applications involve applying classification algorithms to remotely sensed images to determine information about a specific region on the Earth’s surface. These images are very useful sources of geographical data commonly used to classify land cover, analyze crop conditions, assess mineral and petroleum deposits and quantify urban growth. In this paper, we propose a semantic supervised clustering approach to classify multispectral information in satellite images. We use the maximum likelihood method to generate the clustering. In addition, we complement the analysis applying spatial semantics to determine the training sites and refine the classification. The approach considers the a priori knowledge of the remotely sensed images to define the classes related to the geographic environment. In this case, the properties and relations that involve the geo-image define the spatial semantics; these features are used to determine the training data sites. The method attempts to improve the supervised clustering, adding the intrinsic semantics of multispectral satellite images in order to establish the classes that involve the analysis with more precision.

Towards a Methodology to Conceptualize the Geographic Domain

To date, there are different ontologies for many domains and applications. Users can access them in order to share information, reuse knowledge and integrate data sources for several purposes such as semantic web, data warehouse, e-learning, e-commerce, knowledge representation and so on. Ontology engineering is rapidly becoming a mature discipline, having produced tools and methodologies for building and managing ontologies. However, even with a clearly defined engineering methodology, building a large ontology remains a challenging, time-consuming and error-prone task, since it forces ontology builders to conceptualize their expert knowledge explicitly and to re-organize it in typical ontological categories such as concepts, properties and axioms. An approach to conceptualizing the geographic domain is described. It is oriented to formalize the geographic domain conceptualization according to specifications from the INEGI. The main goal is to provide semantic and ontological descriptions, which represent the properties and relations that describe the behavior of geographic objects by means of concepts. GEONTO-MET is focused on developing geographic application ontologies for the sharing and integrating of geospatial information.

Retrieving Geospatial Information into a Web-Mapping Application Using Geospatial Ontologies

Many types of information are geographically referenced and interactive maps provide a natural user interface to such data. However, the process to access and retrieve geospatial data presents several problems related to heterogeneity and interoperability of the geospatial information. Thus, information integration and semantic heterogeneity are not trivial tasks. Therefore, we propose a web-mappingsystem focused on retrieving geospatial information by means of geospatial ontologies and representing this information on the Internet. Moreover, a Multi-Agent System is proposed to deal with the process related to obtain the tourist geo-information, which aids in the information-integration task for several nodes (geographic sites) that are involved in this application. The agent system provides the mechanism to communicate different distributed and heterogeneous Geographic Information Systems and retrieves the data by means of GML description. Also, this paper proposes an interoperability approach based on geospatial ontologies matching that is performed by the Multi-Agent System in each node considered in the application. The retrieval mechanism is based on encoding the information in a GML description to link each geospatial data with a concept of the ontologies that have been proposed.

Knowledge-Based Method to Recognize Objects in Geo-Images

We present an approach to color image segmentation by applying it to recognition and vectorization of geo-images (satellite, cartographic) using knowledge-based learning and self-learning system. This approach exploits the user’s experience providing the knowledge domain in the form of the pre-scribed feature-attribute set. That is a simultaneous segmentation-recognition system when segmented geographical objects of interest (alphanumeric, punctual, linear, and area) are labeled by the system in same, but are different for each type of objects, gray-level values. We exchange the source image by a number of simplified images (composites). Every composite is associated with certain image feature. Some of the composites that contain the objects of inter-est are used in the following object detection-recognition by means of association to the segmented objects corresponding “names” from the user-defined subject domain. The specification of features and object names associated with perspective composite representations is regarded as a type of knowledge domain, which allows automatic or interactive system’s learning. Additionally, we describe the fine-to-coarse scale method of the raster-to-vector conversion in which the “knowledge” of cartographic patterns into small-scale map aids in recognizing the corresponding patterns into large-scale map of the same territory. The results of gray-level and color image segmentation-recognition-vectorization are shown.