Spatial Query
The Spatial Query module
only allows query operations on geo-objects that might be associated with 2D
graphic representations (points, lines and regions). Since geo-objects
algebra operations may involve spatial restrictions, it would be essential
characterize the spatial relationships that may be divided in (Güting,
1994):
-
topological relationships between two entities are defined in
terms of intersections, inside, adjacent, invariant under translation,
rotation and scale;
-
metric relationships: distance operations supposing the existence
of a metric space (for instance,
distance < 100);
-
directional relationships: above, below, north of, south of etc.
The definition of a
minimum set of operators is a very controversial issue in the literature:
-
Freeman (1975) defines a set of 12 operators: near, far,
touching, adjacent, left of, right of, inside, outside, above, below,
upon, and beneath
-
Egenhofer (1994) uses: DC (disconnection), EQ (equality), PO
(partial overlap), EC (external connection), TTP (tangential proper
part), and NTTP (non tangential proper part).
Since there are many
different proposals, it is very important to formalize the spatial
relationships when we are dealing with geo-objects algebra for a well defined
set of geographic objects (simple cases of regions with no holes and
continuous lines).
We will be using the
terms proposed by Clementi et al. (1993) for topological relationships
analysis between point-line-area
elements: Touch, Intersection, Inside, Cross, and Disjoint.
We say that a set of
points w1 touches another
set w2 if the only thing in common between them is inside the union of their
borders, as shown on the figures below:

Figure 1 - Examples of
the topological relationship touch: two areas (a, b), two lines (c, d), line
and area (e, f, g), a point and a line (h) and a point and an area (i). Adapted
from Clementini et al. (1993).
We say that a set of
points is Inside
another w2 set when the intersection of the two sets of points is the w1 set
itself (see Figure 2 below). In this case we can also say that w2 contains w1. If, besides that,
the set of points of w2 is inside the w1 set, we say that w1 set is equal to w2 set.

Figure 2 - Examples
of the Inside
relationship.
We say that the w1 line intersects the w2 line when the
intersection happens in an internal point to both of them (notice that the
intersection of their boundary points would be defined as Touch); similarly,
a line intersects
an area if the line is partially inside and outside this area.
We say that a set of
points w1 covers a
set w2, or, equivalently that a set of points w2 is covered by w1, when the
results of their intersection is a figure with the same dimensions of both. This
relationship is only valid for homogeneous elements.
Finally, two set of
points are disjoint if
their intersection is empty.
See the examples below:

Figure 3 - Examples of relationships:
1. Cross between two lines(a),line
and area
(b, c).
2. Intersection between two areas(d),
two lines
(e, f).
3. Disjoint between two areas
(g), line and area (h),
two points (i).
(Adapted from Clementini et al. (1993)).
Relationships based on topology of geographic entities allow queries such as
"list all public schools from Jardim Satélite district".
In case of directional relationship, SPRING implements the following criteria:
North, South, East, West,
Northwest, Northeast, Southeast, Southwest and Center.
The
directional relationship
is represented by a 3x3 matrix where the central element corresponds to the
bounding box of the base element for comparison. The other cells correspond to
the neighbor positions of the same dimension on the basic cardinal directions:
[1, 1] - Northwest; [1, 2] - North; [1, 3] - Northeast;
[2, 1] - West; [2, 2] - Center; [2, 3] - East;
[3, 1] - Southwest; [3, 2] - South and [3, 3] - Southeast.
The analysis is done from the position of the matrix central cell that contains
the base element for comparison. The bounding boxes of the objects that might
intercept the directional cells are used as comparison elements.
This kind of search allows queries such as: "list the schools located on the
northeast of a certain district".
The
metric relationship
implemented in SPRING allows the use of proximity between objects as a search
criterion. It is necessary to fix some boundaries for the distance using the
operators "<" and ">=" and values provided directly on the interface.
The search using metric criterion allows queries such as: "list all schools
located less than 500 m from the highway".
The "Data to be compared"
item on the query interface is used to choose objects collections in cadastral
maps that will be compared with elements from the comparison base. They must
correspond to the features of the cadastral map active in the control panel.
The "Selection Option"
item contains the following options:
All: selects all objects
that are active on the query module.
New: performs a new query
based on the options chosen and active objects.
Add: adds the new
outcome, using all objects, to the set of current selected objects.
Refine: uses a query only
on the current selected objects.
Check below the basic steps for spatial query operation in SPRING.
Spatial Query operation:
-
visualize on the active
screen and the Cadastral model IL where you want to perform the spatial
query and
click on
Query...
-
on the Visualizing
Objects window, click on Edit - Spatial Query...
-
choose the type of
Operation (Topology,
Direction or Metric) and the type of Relationship. For topological operations it is
only possible to select a relationship at a time. For direction
operations several relationships might be selected from the list. In
case the operation is metric, the relationship list is disabled and the
user must define a value in meters for the distance.
-
choose the
Objects category, IL and a Collection (if this is the
case) that will be compared with the active data from the query. For
example, imagine that the objects being queried are they District type and we want to
know which districts are crossed by a certain River. In this case, we must
select the object category River, the information layer where these
objects are represented, and a collection that might be ALL (all objects
will be represented) or any collection that has been previously defined
(
generating collections modules).
-
select a query option (All,
New, Add, Refine - see below).
-
click on Apply. The
outcome is presented on the active screen.
Canceling a query:
-
on the Visualizing
Objects window, click on Edit - Reset Queries
See also:
Database Query
How to control Objects Visualization .
Attributes Query.
Spatial Query
Attributes
Table.
Pointing and Photo Module.
Query
Models.
|