Geographic Data Input


Outline

  1. Problems With Map Input
  2. Map Digitizing
  3. Map Scanning
  4. By-Passing the Map with GIS
  5. Questions:

Problems with Map Input

Scale
A map is basically a reflection of scale on paper. Different map scales allow different levels of detail. Large-scale maps will allow more details that can be accurately displayed. Small-scale maps allow less detail. On small-scale maps, features may need to be moved around to accommodate the desired amount of detail.

Symbolization
As with scale, symbolization is a representation on paper. A problem with symbolization is that it changes between maps of different scales. On a small-scale map, cities may be point data represented by dots. On a large-scale map, cities may be area data represented by shaded areas.

GIS Overlays
Inconsistencies may occur between two maps in an overlay if the source maps from which the two maps are made are of different scales. This becomes especially evident with features such as rivers that also serve as borders.

Map Digitizing

Human Approach
A source map and a piece of grid paper are overlaid on a light table to find the x-y coordinates of features on the map. The data is then used as Teletype input.

Hardware Approach
An x-y coordinate digitizer consists of a large tablet and a puck. The tablet is essentially a table with a wire grid (instead of a grid paper) running through it. The puck is a magnetic device that sends signals that can be detected by the grid in the tablet. The puck is used to determine x-y coordinates of features on the source map. Digitizers generally have .001" resolution and a .003" accuracy, and range in cost from just under $500 to more than $7,000.

Software Approach
Digitizing software takes x-y coordinates input from the puck and tablet and displays them on the computer screen. The data is stored in an arc/node format. For each pair of x-y coordinates input, the user defines whether or not it is a point or a node. In addition, the user can assign attribute data to each point or node. ArcInfo is an example of a software package that has digitizing capabilities.

Human Factors
Digitizing is extremely labor intensive. It leaves a lot of room for operator error, and it is time consuming. U.S.G.S. would limit its operators to 45 minutes of digitizing at one time to cut down on the amount of error. Another problem with this labor-intensive method is that it is expensive in this country since labor costs are relatively high here.

Offshore Digitizing
As a result of the high cost of labor in the U.S., most maps are sent overseas to be digitized in countries with low labor prices. Counties commonly associated with offshore digitizing include India and Taiwan. In some of these countries, labor so cheap that maps are digitized two different times, by two different operators. The differences between the two digitized maps can then be taken into account before the final map is sent out. The U.S. Government however does now forbid any of its contracts to go to foreign countries for map digitizing.

Map Scanning

Hardware
The scanners used have resolutions in the several thousandths on and inch range. They are typically of the drum type in which the map is secured to the outside of a large drum that is rotated as the scanning head stays in place. The drawback is that these scanners are extremely expensive.

Software
The main function of the software used in map scanning is to perform rater to vector conversion of the scanned map features. The lines on scanned maps are several pixels wide due to the high resolution of the scanners. To vectorize the map, lines are skelotonized by the software. To do this the software finds the middle pixels that represent the line. Once this is done, attribute data can then be attached to the vectorized map.

Service Companies
Because of the high amount of capital involved with map scanning equipment, Service Companies have set up to contract the jobs. When these maps are made, they are usually composited from several individual feature maps (i.e. rivers, roads, towns, etc.). These service companies are equipment intensive, so they have very few employees. The U.S. Government uses these contractors.

By-Passing the Map with GPS

GPS Satellite Constellation
The GPS Constellation consists of 24 satellite that were built by Rockwell International and launched by the U.S. Air Force. 21 of these satellites are in primary operation, while 3 serve as spares. The satellites orbit the earth twice a day at an altitude of 10,900 miles. This high altitude assures that the orbits remain stable and precise.

Measuring Distance to Satellites
The distance of GPS receivers to orbiting satellites is measured by timing how long the signal takes to travel between. The radio signal travels at the speed of light. Since this speed is precisely known, is multiplied by the signals travel time to find the distance.

Trigonometry Gives Position
By measuring known distances to two satellites from a spherical surface (the Earth), you can narrow your location down to a circle on the sphere. By measuring known distances to three satellites from a spherical surface, you can narrow your location further to two points on the sphere. With introduction of a third satellite of known distance, you can figure out which of the two points is your location.

Accuracy
For civilian use, the Air Force degrades the signal to an accuracy of 60m. This low level of accuracy can be overcome by using differential GPS.

Differential GPS

Logging Data
To log data, the GPS user simply moves along a geographic feature as the receiver continuously logs position. Generally, while the user is collecting data in the field, the base station is simultaneously collecting data to do differential correction during the data processing phase.

Real-time GPS
This form of GPS data logging performs differential correction of information as the receiver moves to create a map.


Submitted by Scott A. Carson on February 6, 1998