Recent growth of the geospatial information on the web has made it possible to easily access various maps and orthoimagery. By integrating these maps and imagery, we can create intelligent images that combine the visual appeal and accuracy of imagery with the detailed attribution information often contained in diverse maps. However, accurately integrating maps and imagery from different data sources remains a challenging task. This is because spatial data obtained from various data sources may have different projections and different accuracy levels. Most of the existing algorithms only deal with vector to vector spatial data integration or require human intervention to accomplish imagery to map conflation. In this paper, we describe an information integration approach that utilizes common vector datasets as "glue " to automatically conflate imagery with street maps. We present efficient techniques to automatically extract road intersections from imagery and maps as control points. We also describe a specialized point pattern matching algorithm to align the two point sets and conflation techniques to align the imagery with maps. We show that these automatic conflation techniques can automatically and accurately align maps with images of the same area. In particular, using the approach described in this paper, our system automatically aligns a set of TIGER maps for an area in El Segundo, CA to the corresponding orthoimagery with an average error of 8.35 meters per pixel. This is a significant improvement considering that simply combining the TIGER maps with the corresponding imagery based on geographic coordinates provided by the sources results in error of 27 meters per pixel.

Abstract Recent growth of the geospatial information on the web has made it possible to easily access a wide variety of spatial data. The ability to combine various sets of geospatial data into a single composite dataset has been one of central issues of modern geographic information processing. By conflating diverse spatial datasets, one can support a rich set of queries that could have not been answered given any of these sets in isolation. However, automatically conflating geospatial data from different data sources remains a challenging task. This is because geospatial data obtained from various data sources may have different projections, different accuracy levels and different formats (e.g., raster or vector format), thus resulting in various positional inconsistencies. Most of the existing algorithms only deal with vector to vector data conflation or require human intervention to accomplish vector data to imagery conflation. In this paper, we describe a novel geospatial data fusion approach, named AMS-Conflation, which achieves automatic vector to imagery conflation. We describe an efficient technique to automatically generate control point pairs from the orthoimagery and vector data by exploiting the information from the

Abstract—In this paper, a new two-step approach (detecting and pruning) for automatic extraction of road networks from aerial images is presented. The road detection step is based on shape classification of a local homogeneous region around a pixel. The local homogeneous region is enclosed by a polygon, called the footprint of the pixel. This step involves detecting road footprints, tracking roads, and growing a road tree. We use a spoke wheel operator to obtain the road footprint. We propose an automatic road seeding method based on rectangular approximations to road footprints and a toe-finding algorithm to classify footprints for growing a road tree. The road tree pruning step makes use of a Bayes decision model based on the area-to-perimeter ratio (the A/P ratio) of the footprint to prune the paths that leak into the surroundings. We introduce a lognormal distribution to characterize the conditional probability of A/P ratios of the footprints in the road tree and present an automatic method to estimate the parameters that are related to the Bayes decision model. Results are presented for various aerial images. Evaluation of the extracted road networks using representative aerial images shows that the completeness of our road tracker ranges from 84 % to 94%, correctness is above 81%, and quality is from 82 % to 92%. Index Terms—Bayes decision rule, road extraction, road footprint, road tracking, road tree pruning. I.

Recent growth of the geo-spatial information on the web has made it possible to easily access various and high quality geo-spatial datasets, such as road networks and high resolution imagery. Although there exist efficient methods to locate road intersections from road networks for route planning, there are few research activities on detecting road intersections from orthoimagery. Detected road intersections on imagery can be utilized for conflation, cityplanning and other GIS-related applications. In this paper, we describe an approach to automatically and accurately identifying road intersections from high resolution color orthoimagery. We exploit image metadata as well as the color of imagery to classify the image pixels as on-road/off-road. Using these chromatically classified image pixels as input, we locate intersections on the images by utilizing the knowledge inferred from the road network.

Many Geographic Information System (GIS) applications require the conversion of an address to geographic coordinates. This process is called geocoding. The traditional geocoding method uses a street vector data source, such as, Tigerlines, to obtain address range and coordinates of the street segment on which the given address is located. Next, an approximation technique is used to estimate the location of the given address using the address range of the selected street segment. However, this provides inaccurate results since the approximation assumes that properties exist at all possible addresses and all properties are of equal size. To address the inaccuracy of the traditional geocoding approach, we propose two new methods for geocoding using additional online data sources. The first method, the uniform-lotsize method, uses the number of addresses/lots present on the street segment to approximate the location of an address. The second method, the actual-lot-size method, takes into consideration the lot sizes on the street segment and the orientation of the lots as well. Moreover, we describe an implementation of these methods using an information mediator to obtain information about actual number of lots and sizes of the lots on the streets from various property tax web sites. We geocoded an area covering 13 blocks (267 addresses) using all three methods. Our evaluation shows that the traditional method results in an average error of 36.85 meters, while the uniform-lotsize and the actual-lot-size methods result in the average error of 7.87 meters and 1.63 meters, respectively.

Abstract. The traditional way of representing motion in 3D space-time uses a trajectory, i.e. a sequence of (x,y,t) points. Such a trajectory may be produced by periodic sampling of a Global Positioning System (GPS) receiver. The are two problems with this representation of motion. First, imprecision due to errors (e.g. GPS receivers often produce off-the-road locations), and second, space complexity due to a large number of samplings. We examine an alternative representation, called a nonmaterialized trajectory, which addresses both problems by taking advantage of the a priori knowledge that the motion occurs on a transport network. 1

Incomplete knowledge about the quality of data is a fundamental issue that needs to be addressed by the spatial information community over the next decade. Users of spatial data must be able to easily ascertain the quality of their information and its ability to meet their requirements. They also need to know where quality varies throughout a dataset; the degree of uncertainty that is associated with any of their derived information products; and, for non-experts in particular, there needs to be an improved way of communicating data quality—especially in the context of web-based metadata. To help address these 'next-generation ' research issues, a project has been initiated by the CRC for Spatial Information which aims to design, develop and test new modelling and visualisation solutions for communicating spatial data quality and uncertainty to future users of spatial data.

the information sharing from different geographic databases, also known as “GIS data interoperability”. This is a huge problem, involving several aspects. Among them, we decided to concentrate an a geometric conflation of different maps since maps coming from different sources often show significant geometric differences. In order to solve, or at least mitigate this problem, a number of different approaches were proposed in the past years. Particularly referring to the specific case of vector maps, almost at the same scale, recently the authors presented a possible solution based on the automatic detection of homologous points on different maps and a further transformation based on multiresolution spline functions. To extend the proposed approach also to the case of raster maps, the possibility of using the point matching approach on vectorized maps obtained from raster ones was investigated. In order to avoid duplicating the effort of developing software and, at the same time, of taking advantage of existing free code, the first attempt, here presented, focuses on the application of our method on vectorized maps obtained by using the software Ras2Vec. 1.

Vast amount of geospatial datasets are now available through numerous public and private organizations. These datasets usually cover different areas, have different accuracy and level of details, and are usually provided in the vector data format, where the latitude and longitude of each object is clearly specified. However, there are scenarios in which the spatial attributes of the objects are intentionally transformed to a different, and usually unknown, (alien) system. Moreover, it is possible that the datasets were generated from a legacy system or are represented in a native coordinate system. An example of this scenario is when a very accurate vector data representing the road network of a portion of a country is obtained with unknown coordinate. In this paper, we propose a solution that can efficiently and accurately find the area that is covered by this vector data simply by matching it with the (possibly inaccurate and abstract) data with known geocoordinates. In particular, we focus on vector datasets that represent road networks and our approach identifies the exact location of the vector dataset of alien system by comparing the distribution of the detected road intersection points between two datasets. Our experiment results show that our technique can match road vector datasets that are

In this paper, we will focus on the semantic heterogeneity problem as one of the main challenges in current Spatial Data Infrastructures (SDI). We first report the state of the art of the current approaches and proposed solutions to overcome these semantic heterogeneity problem focusing both on geo-services integration and geo-data integration. We then propose a general framework to integrate different application ontologies used by geographic information providers. The proposed framework is based on a novel view of the semantics of Web services coordination, implemented by using the Lightweight Coordination Calculus (LCC) language. In this approach, the services share explicit knowledge of the interactions in which they are engaged and these models of interaction are used operationally as the anchor for describing the semantics of the interaction. To discover and integrate geo-services and related geo-metadata, semantic matching need to be performed. The proposed framework explore the applicability of the methodology and the solution proposed by the S-Match system, a semantic matching system built at the University of Trento on top of a number of approaches both at the element level and at the structural level matching.