Home > Georeferencing
3.0. Georeferencing
A crucial element of any parcel mapping project is registering a parcel map
with the correct real world coordinates. This procedure is called
georeferencing. If the parcel maps are not georeferenced, no other information
can be displayed over or positioned under the parcel map coverage.
Georeferencing could be done either before or after the process of parcel data
conversion. There are two groups of methods that could be used for this purpose
depending on source materials available in each particular case:
1) Registration to a coordinate system; and
2) Registration to a base map
3.1. Registration to a Coordinate System
3.1.1. Use of Existing
Coordinates
The first method of digital parcel map registration is to use the known
coordinates. Currently, most of the towns in Massachusetts have their parcel
maps in the form of large blueprints that were originally either hand drafted
or done in Auto CAD. Some of the towns may even have coordinates displayed in
the corners of individual parcel maps. It is likely that these coordinates are
in Massachusetts State Plane Coordinate System, North American Datum (NAD) 1927
or NAD 83. If they are in NAD 27 Datum, after parcel map registration it is
necessary to transform the map to NAD 83, which is a standard for
Massachusetts. To use the known coordinates is the easiest way to register the
map. If the known coordinates are used, maps line up precisely.
3.1.2. Use of Artificial
Coordinate System
Unfortunately, not all towns possess tax maps registered to a known coordinate
system. In such cases, artificial coordinate systems are frequently used. One
way to create such a coordinate system is to assign the lowest corner a (0,0)
coordinate and then calculate coordinates for all other corners of each parcel
map on the basis of the distance from the original (0,0) point.
Figure 3.1: Artificial coordinate system construction
A parcel map created by this method will not be registered to any real world coordinate system but the distance between any parcels on the map will be accurate relative to each other. Afterwards it will be possible to register this map to some existing coordinate system or to a base map.
3.1.3. Use of GPS
Technology
Another way to create the coordinate system for the town is to use Global
Positioning System (GPS) technology.
GPS is a "navigation and positioning system developed by the U.S
Department of Defense for use by U.S and Allied military forces" (Land
& Mapping Services, p. 1). However, GPS very quickly expanded to
non-military applications and now it is used by water and air navigators,
geographic surveyors, mappers, geologists, archeologists, forecasters, and many
others. Through latitude/longitude or some other coordinate system, GPS can
determine the location of the GPS receiver.
GPS works in the following way. There are 24 satellites 11,000 miles above the Earth that transmit orbital position information, current time, and error corrections. In order to determine the location taken at GPS receiver position, it is necessary to receive signals from four satellites. The GPS receiver interprets these signals and determines the distance between the receiver and each satellite. The distance is calculated by multiplying the speed of the signal by the amount of time it took the signal to travel from the satellite to the receiver. Now for each satellite the receiver can plot a radius, with the satellite as a center. The point where all radii intersect, is the GPS receiver's location. (Red Horse Technologies, 1998, 1-2).
GPS receivers differ among each other
depending on the accuracy they produce and their cost. It is possible to
distinguish three types of GPS devices depending on their use:
1. The most basic GPS receiver - "handheld" unit. Provides accuracy
of several hundred feet.
2. Professional, "mapping" grade receivers. Provide accuracy of
several feet.
3. "Survey grade" or Phase based receivers. Provide accuracy of
several inches.
Source: Land & Mapping Services, p. 2-3.
There are several factors that can affect
accuracy of the GPS readings:
1. Equipment error.
2. View of the sky may be obstructed by trees and other objects.
3. Signals may bounce off the buildings, mountains, etc.
4. The Department of Defense may turn on "selective availability",
which is an intentional clock error introduced into the timing signals.
One of the GPS applications is parcel mapping. GPS may be used to determine coordinates of easily identifiable positions on a parcel map (such as roads intersection, property corner, etc.). In order to register a parcel map to real world coordinates it is necessary to have at least four coordinates. But in order to achieve greater accuracy usually it is better to have more than four coordinates. Another issue is the placement of the coordinates - if all four or more coordinates are concentrated in one corner of the map, it will be possible to register the map, but all other corners will be skewed and accuracy of the map will be very low. The choice of the number of coordinates in each case depends on the shape and complexity of each particular parcel map. In general, however, the more coordinates are used, the more accurate the final map will be.
There are certain benefits of this method of
parcel maps registration:
1. It is inexpensive
2. It is easy to use
3. It is not time consuming relative to other methods
4. It is not labor intensive
5. It is accurate
3.2. Registration to a Base Map
A base map is "…the graphic representation at a specified scale of selected fundamental map information … (that) provides a primary medium by which the locations of cadastral parcels can be related to … major natural and man-made features such as bodies of water, roads, buildings and fences…" (National Research Council, 1983 in Goodwin 1994, 16). Several types of sources may be used as a base map. This includes orthophotographs, and already existing maps, such as road network maps, and topographic maps.
3.2.1. Use of Orthophotographs
as a Base Map
An orthophoto is an aerial photograph that is geometrically correct. On an
orthophoto inaccuracies due to displacement, distortion, aircraft movement, and
camera tilt are removed via the process of orthorectification, rendering surface
features with great accuracy. A normal aerial photograph does not show all
features in their correct location (see Figure 3.2)
Figure 3.2:
Displacement in an aerial photograph
Source: OGRIP, p. 3
Orthorectification removes the distortion as shown in Figure 3.3.
Figure 3.3:
Orthorectification
Source: OGRIP, p. 3
One of the uses of orthophotos is in parcel mapping. "Orthophotos make excellent bases for cadastral and tax mapping" because they "…include faint and often intermittent linear features, such as lot boundaries, cut lines through woods, field lines, fences, etc., that are helpful when locating original property lines" (Thrower and Jensen 1976 in Goodwin 1994, 20).
The most important characteristics of orthophotographs that affect their use are scale, accuracy, and resolution. In the case of the hard copy orthophotos, final representation scale determines resolution and accuracy of the displayed features. In the case of digital orthophoto, scale can be changed, but resolution (pixel size) limits what features can be seen when scale is increased. In general, the larger the scale of the orthophoto, the better its resolution and accuracy (OGRIP, p. 4).
|
Scale |
Ground Resolution (Pixel Size) in Feet |
Accuracy in Feet |
Parcel Dimension in Sq. Feet |
Cost per square mile |
|
1:4,800 (1"=400') |
1-2 |
10-15 |
208.7 |
$125-$165 |
|
1:2,400 (1"=200') |
1 |
3-5 |
60 |
$260-$350 |
|
1:1,200 (1"=100') |
0.5-1 |
1-3 |
30 |
$750-$1,000 |
Figure3.4: Orthophoto scale, resolution,
accuracy, corresponding parcel size and cost
Source: OGRIP, p. 11, 16.
After scale and cost, timing of flights is
also important. For parcel mapping purposes "leaf off" aerial
photography acquired in the fall or spring (before or after snow cover) during
the middle of the day is preferred (OGRIP, 8). As for the choice between black
and white vs. color imagery, for most parcel mapping situations, black and
white data should suffice (OGRIP, 9).
Both digital and hardcopy orthophotographs can be used as a base map.
3.2.2. Hardcopy
Orthophotos
Hardcopy orthophotos can be made either on paper or on Mylar. Mylar is a more
stable medium as it is not affected by moisture or temperature that could cause
stretching or shrinkage of the paper maps. First, it is necessary to recompile
the parcel boundaries to the orthophoto. One way to do it is to use a
cartographic instrument called a Zoom Transfer Scope (ZTS) or similar type of
device. ZTS uses lenses, mirrors, and light to change the scale of one map (or
photograph or image) so that it would be possible to superimpose it on another
map (in this case orthophotograph) (Goodwin 1994, 44). When the transfer of the
parcel boundaries to the orthophoto is complete, it is necessary to
georeference the data. The coordinate system associated with the orthophoto can
automatically be used for this purpose. If no coordinate system is associated
with the orthophoto, it is necessary to choose at least four points easily
recognizable both on the orthophoto and on the ground for each orthophoto and
obtain coordinates for these points by use of GPS or some other method.
3.2.3. Digital
Orthophotos
Digital orthophotos have several advantages over hardcopy ones. First of all,
they are not subject to shrinkage and stretching thus giving a better accuracy.
Second, they are already georeferenced. Thus the problem of obtaining
coordinates for registration points doesn't exist. Third, they are easy to
store. Finally, they offer an opportunity to use different methods of parcel
conversion. One of the methods is heads-up digitizing directly from the
orthophoto. In this case, assuming the orthophoto is already registered to a
correct coordinate system, these coordinates could be used for the parcel
coverage. Another method is scanning tax maps. In this case, an orthophoto may
be used as a source of ground control points also common to the scanned images.
Using these coordinates, scanned images may then be converted to the same
coordinate system as the orthophoto. After that either heads-up digitizing
procedure or vectorizing could be used (Torres, 4-5).
3.2.3.1. Example of
Digital Orthophoto
Figure 3.4 gives an example of the 1-meter pixel resolution digital
orthophotograph. This orthophoto is a portion of the bigger image downloaded
from MassGIS web site: http://www.state.ma.us/mgis/ftpintro.htm
Figure 3.4: Digital Orthophoto.
3.2.4 Sources of
Orthophotography in Massachusetts
In 1989 Massachusetts Department of Environmental Protection made a decision to
produce orthophotoquads at a scale of 1:5,000 (1"=417') (Goodwin 1994, 18
and MassGIS). This decision was based upon the state's interest in very
accurate wetland mapping as well as providing highest resolution base maps for
a wide variety of larger-scale uses. At this time only around 35% of the state
is covered by this type of base map. Statewide coverage is expected to be
completed in the next three years. The maximum displacement of well defined
features on these orthophotos is less than 5 meters. Each pixel in the digital
orthophoto images represents 0.5 meters on the ground. Each image has been
resampled at 1, 2, and 5 meter resolutions. The status map showing communities
for which digital orthophotography has already been completed could be accessed
at MassGIS web site at http://www.state.ma.us/mgis/st_oq.htm.
MassGIS contact information:
Julie Sweitzer, Program Coordinator
20 Somerset St. - 3rd Floor
Boston, MA 02108
Phone: (617) 727-5227 x323 Fax: (617) 227-7045
E-mail:julie.sweitzer@state.ma.us
Other contacts:
Data and map orders (status and information): Greg Scott, ext. 327
MassGIS Data Viewer: Aleda Freeman, ext. 326
MassGIS Database/Orthophotos: Michael Trust, ext. 322
Another possible source of existing orthophotography could be local utility companies or U.S. Geological Survey (USGS): http://www.usgs.gov/. If none of these sources has orthophotography of required resolution for the particular area, it may be necessary to schedule a flight and produce a new set of orthophotographs. This will, however, add considerable cost to any project.
| Next Section: Core Attributes | Home |