University of Maryland Eastern Shore Intersect and Union Discussion

1) Which different types of outputs are possible (points, lines, polygons) when performing intersect and union?

2) What is the most important difference between a spatial join and a map overlay?

GIS580: Geospatial Intelligence
Lecture 7
Professor, Dr. Sergei Andronikov
Dean of Academics
Measurements
• Location
• SIZE – length, areas, perimeters.
• Ratios.
• Length:Width. Perimeters:Areas
• Points – …….. Dimensions;
• Lines – ……….. Areas – …………….
Measurements
• Polygon Length.
• In Raster: lack techniques.
• In Vector: the solution is of calculating the lengths of each pair of opposing
polygon vertices.
• Orientation. Relationship between the major and minor axes. A ratio of short
axis to long.
• Polygon Area.
• Area gives a quantitative measure of the amount of each classified areal
feature.
• In Raster: grids that share common attributes. Just read the number, no
calculations. Or reclassify.
• In Vector: polygon is selected based on coordinate locations or by attribute
values in attribute tables.
• A measure of the complexity – perimeter/area ratio.
Measurements
• Polygon Shape
• 1. Spatial Integrity – a measure of the amount of perforation.
• 2. Boundary configuration.
• A measure of convexity and concavity.
• In Vector:
•
CI = kP/A
• In Raster:
•
CI = P/  #cells
• EDGINESS. The lowest amount of edge possible.
• Roving window, filter.
• Edge enhancement, smoothing.
• The more Values, the less edginess.
Raster Measure of Edginess
• Moving 3×3 Filter
Edginess index = 2
1
1
1
0
1
0
1
0
0
0
0
0
1
1
0
1
1 1
Edginess index = 8
Classification
• An age-old problem.
Classification
• All classifications have a target audience or end users in mind.
• The more closely classification is made to fit user needs, the more
useful it becomes.
Reclassification
• Based on locational attributes
• In Raster:
• You must change the numbers that represent attributes,
attribute code;
• Then change the legend for the new coverage.
• In Vector:
• To change both entities and attributes
• First, a line dissolve operation.
• Then, data aggregation.
• Overlay operation.
• It limits us to the attributes within each object/feature
Neighborhood functions
• To classify ….flying over the area.
• Neighborhood functions – based on the idea to
characterize an object as part of a larger
neighborhood of objects based on some SHARED
attributes.
• Static neighborhood functions – the analysis takes
place all at once for the selected target area.
• Roving window neighborhood functions (in
raster) – the analysis takes place within the
framework of a window that moves across the
coverage.
Windowing functions
• FILTERS
• Used to reclassify whole raster coverages based in a wide range of
alternative filter values and combination of values.
• Frequently in image processing.
• HIGH PASS FILTER – to separate out detail in raster coverage that
may be obscured by nearby grid cells with relatively similar values.
• To enhance the short-range spatial properties. E.g. To obtain
boundaries. to find small topographic ridges in raster.
• LOW PASS FILTER (SMOOTHING) FILTER. The value at the center is
a simple arithmetic average. Produce smoother image. Also
compute max. min, mode, St.Dev.,
• Original Surface – Low-Pass Filter = High-Pass Image.
Polygonal Neighborhood
• SUPPOSE: Identity of a region’s polygons within a certain neighborhood or
distance.
• E.g. The spread of innovative farm practices. Neighborhoods of No-till
agriculture.
• A. Retrieve a coverage showing ONLY locations in which no-till agriculture.
B. Select a distance value of where limitations are occur. The result – small,
medium, large group of fields that seem to indicate the sharing of the new
practice.
• Each group can be later reclassified by measuring the size.
• BUT we are LESS interested in groups of IDENTICAL polygons, MORE in
defining the similarities WITHIN a neighborhood distance.
• E.g. the average age in a specified region. Max Number of crimes for a given
year. Min Value for the houses.
• Select + Average (MAX, MIN) = NEW coverage.
Polygonal Neighborhood
• Jungle + Running Water (fish) +200 Square mile of territory
• Researchers: 1 part stream corridor —- 10 parts jungles.
• 1. Identify the ratio between target points along the stream from stream coverage
an the amount of surrounding forest from the vegetation coverage
• 2. Reclassify the vegetation coverage into stream corridor and forest vegetation.
• 3. Using the neighborhood function that computes the ratio of the amount of
neighboring area having the same category as a target, IDENTIFY the
neighborhoods that have at least 1/10 stream corridor.
• Finally, to consider size – > 200 sq.miles.
• Using MEASURING distances across polygonal objects, add the analysis that
locates and classifies regions based on a value of narrowness
BUFFERS
• A BUFFER is a polygon created through reclassification
at a specified distance from a point, line, or area.
• It is based on a position: location, shape, and orientation
of the object.
• 1. Arbitrary Buffers. Zones are guessed.
• 2. Causative Buffer. Based on a priori knowledge
(difference in soil permeability)
• 3. Measurable Buffer. Accurate measurement.
• 4. Mandated Buffer. The value is measurable, selection
over other values is arbitrary (how close to a fire hydrant
to park).
BUFFERS
• Buffering
Spatial Arrangements
• Rather than looking at the amount of space occupied by an object, we
will look at overall arrangement of objects in space.
Comparing Variables
• OVERLAY requires BOTH graphic and attribute comparisons.
Social Factors
Biodiversity
Engineering
Land Use
Environmental
Considerations
OVERLAY OPERATIONS
• 1. POINT-IN-POLYGON operation
• E.g. vegetation map>>>map of point locations of cowbirds>>>resultant
map.
• 2. LINE-IN-POLYGON operation
• E.g. transportation network and urban expansion.
Types of Polygon Overlay
• 1. Boolean or logic (OR, AND, etc.)
• 2. Mathematically based (subtraction, division, squares, min, max,
average, etc.)
• 3. Selective (rules-of-combination)
• E.g. If moderate or less soil limitations for housing AND
moderate to no soil limitations for agriculture OR greater than
200 meters from an agricultural site THEN minimize coverage
values ELSE maximize coverage values
• 4. Identify overlay
• You can isolate any of the numbers and preserve all their
attribute information.
Polygon Overlay
• OVERLAY is powerful BUT mindless…
• Overlay is FAR MORE powerful when
combined with other methods of spatial analysis
CONGRATULATIONS!!!!!
• You came to the end of your
FIRST-EVER GIS COURSE!
• Now, you can have some rest & fun!
• http://www.wimp.com/fireworksshow/