in
T
y
al
SPATIAL DATA REVISION: TOWARD AN INTEGRATED SOLUTION USING NEW TECHNOLOGIES
J. Raul Ramirez, Ph.D.
The Ohio State University Center For Mapping, USA
Commission IV, Working Group 4
KEY WORDS: Geomatics/GIS, Cartography, Revision, Model, Digital
ABSTRACT:
Geographic Information System (GIS) is one of the fastest growing technologies in the world. Its ability to analyze and provide answers
to many spatial problems is impressive. Spatial data are the backbone of GIS analysis, but only current and accurate spatial data can
provide the appropriate framework for successful use of GIS technology. Out-of-date and/or inaccurate spatial data could contaminate
GIS results. Therefore, there is a great need for cost-efficient spatial data revision and quality control methods.
The Center for Mapping is involved in a large spatial data conversion effort, the Generating Information from Scanning Ohio Maps
(GISOM) project. The GISOM project is converting to computer-readable form all 793 7.5-minute quadrangle maps produced by the
U.S. Geological Survey (USGS) covering the State of Ohio. The average age of these maps is twenty years. The Center recognizes the
capital importance of revising these data in order to have the base needed by the State of Ohio to make full use of GIS technology.
The Center for Mapping is developing a conceptual framework for spatial data collection (including conversion) and revision and
investigating how to integrate local collection and revision of spatial data with modern technologies such as: mobile mapping systems,
spatial data conflation, digital photogrammetry, digital terrain models, etc. This paper presents a summary of the conceptual framework
of spatial data collection and revision and describes the GISOM project and the latest results of our spatial data revision research.
Finally, future research directions are discussed.
1. INTRODUCTION
Geographic Information System (GIS) is a growing technology.
More and more users are finding its capability to analyze and
query geographic information of incredible help in understanding
the environment and making better decisions. Today, the major
limitation to the use of GIS technology is the limited availability
of digital spatial data. As an example, in 1991, less than 5% of
the maps of the United States of America at scale 1:24,000 (the
primary topographic map series) had computer-compatible
representation.
Collection of computer-compatible spatial data in vector format
is a costly and time-consuming process. The two major
approaches are: (1) digital spatial data collection from the terrain
(using remote sensing or mobile mapping systems techniques or
a combined approach) and (2) conversion of existing maps into
digital representations. Both approaches are operator-intensive.
Mobile mapping systems currently offer the highest degree of
automation and accuracy for the collection of road and railroad
data. However, the collection of all the data included in a general
purpose spatial database requires a lot of human intervention.
It took almost forty years and hundreds million of dollars for the
USGS to complete the analog coverage of the United States at
scale 1:24,000. It can be argued that a country the size of the
United States is impossible to remap (in digital form) in entirety
because of cost and time constraints. The most cost-efficient
solution to generate the digital spatial data needed by GIS in a
country the size of the United States is, perhaps, a combination
of remapping selected areas, conversion of existing analog maps,
and revision of digital spatial data. This paper introduces the
677
topic of spatial data collection and revision, describes briefly the
map conversion effort at the Center for Mapping, describes the
status of our research in this area, and finally, discusses future
research directions.
2. SPATIAL DATA REVISION
Spatial data revision is defined by Ramirez (1996) as correcting,
updating, and improving the content of existing data to obtain a
current representation of the terrain, in agreement with a
predefined purpose. The revision effort at the Center for
Mapping is directed toward the DLG-3 files generated by the
GISOM project.
Conventionally, spatial data revision requires the use of current
aerial photographs and manual identification and compiling of all
the changes on the terrain. With the increasing use of
computer-based methods, partial revisions are possible today.
There is not a universally accepted spatial data revision method.
In general, spatial data revision is agency dependent. For the
purpose of providing an example of revision, a brief description
of the USGS’s method follows. In agreement with Thompson
(1987) and others, map revision is divided into four major tasks:
total revision, partial revision, photorevision, and photo-
inspection. Total revision is the “correction of all deficiencies in
planimetry and relief features” and it is the only type of revision
that keeps a consistent terrain representation. Photo-inspection is
the process of “comparing the latest published map to recent
aerial photographs to determine both the need for revision and
the extent of the changes.”
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996