| Some web search engines index only the first
part of long documents. In order to have this document indexed
fully, it was divided into shorter pieces. |
1
2
3
4
5
6
7
8
9
|
| If you have
arrived here from a web search, go to the full document ... |
View in PDF
format
* Manifolds.
Background.
TNT
products and some other advanced commercial geospatial analysis systems are
often referred to as 2.5 dimensional (2.5D).
The Z axis for features created, used, and visualized in these systems is
measured in a vertical axis orthogonal to the 2D projection of these features
onto a map plane. Often this plane
is defined at or referenced to mean sea level.
Generally, the available 3D data such as DEMs, bore hole profiles,
seismic profiles, and 3D coordinates for maps fall into this category.
All these 2.5D systems store and use only minimal information about
relationships that occur in the 3rd dimension.
For example, 3D topology is not maintained, in other words the topology
is only 2D and refers only to the projected surface.
While vertical overlapping area features can be defined in a CAD or shape
object, true 3D relationships (for example, caves, concavities, and the general
idea of volumes) are lost in a polygonal topology vector or single raster
object.
The few 3D and 4D geospatial systems that
do exist are specialized, expensive, and/or developmental in nature and focus on
geologic, air-space and atmospheric, oceanic, and outer space related special
applications. Upon careful
consideration you will find that geodata with Z coordinates that are not
measured orthogonally from the projected map plane and/or preserve volumetric
relationships is very hard to collect and, as a result, is very expensive and
sparsely available. Thus, it is
natural that commercially oriented geospatial analysis systems focus upon the
plentiful 2D geodata and on 2.5D when DEMs are available.
Desktop commercial geospatial analysis products, such as TNTmips
and its competitors, are not viable unless they are based upon, and work with,
widely available geodata.
Fortunately there are still many new and
useful 2.5D applications of geospatial analysis similar to the introduction in RV7.0
of the use of manifold surfaces and stereo viewing.
Other recent MicroImages 2.5D introductions included the free TNTsim3D
program for publishing your simulations and the continuing improvements for
higher quality, faster 3D perspective viewing.
Future 2.5D opportunities might include editing 2.5D geodata in stereo;
editing using several geolocked orthogonal views; or sketching in 2D, 3D, and or
stereo views with the results simultaneously added to each. These past, present,
and future developments can exploit the sources of Z-coordinate geodata just
becoming widely available for use with your 2D geodata.
These include more widely available and accurate DEMs ranging from the
global SRTMs to local laser-derived DEMs to collecting accurate elevations with
a GPS unit.
Some applications of 2.5D geodata are
obvious and are the basis for the myriad of
products that create a realistic 3D perspective view or real time
simulation of a 3D surface. These
products strive for speed and realism in games and some use real geodata for
applications like flight simulation or guiding the driver of a car.
However, in many site specific professional geospatial applications, the
level of the integration of these 3D visualization subsystems into the total
system is of even more importance than realism.
For example, can they use the same geodata layers and scripting tools?
Are the 2.5D capabilities and 3D capabilities tightly integrated with 2D
capabilities for visualization, interpretation, and editing, such as in TNTmips
and TNTedit?
Are high quality 3D and acceptable quality stereo created with or without
a DEM to help visualize complex surfaces to create 2.5D geodata relationships,
such as manifolds in 2D views?
Only now in wealthy nations are we
getting into position with our data collection and analysis systems to keep our
geodata systems updated rather than creating static, one-time geodata sets and
their interpretations. And this is
generating a monstrous amount of digital data much of which may never be used.
Obviously these nations have a tremendous amount of effort at hand just
organizing and using these 2D and 2.5D geodata.
Recently a senior project manager for one
of the largest intelligence and homeland security contractors to the U.S. government was overheard to say that it will take 25 to 50 years to actually
integrate the local, state, and national geodata into a comprehensive and
maintained 2.5D geospatial system. It’s
not just a technical issue—spend the money and get it done.
A myriad of local and other laws, institutional barriers, and political
issues must be resolved. Funding
must be available, channeled to the working levels in local, state, and federal
government, and standards truly adopted. There
will have to be unpopular sacrifices of and safeguards for personal freedoms
within our basic democratic framework. For
example, even the limited amount of geospatial data available is already used to
gerrymander election results.
We have plenty of work cut out for us
worldwide just coping with our 2.5D geospatial future. True 3D and 4D systems
are truly long range goals that can be worked toward only if and when sources of
appropriate 3D geodata become available. The
first step will be to move toward 3D geocentric and geodetic coordinate systems
to properly record positions in a geodetic or geocentric system.
MicroImages and other developers are
adding more of these 2.5D applications as demand, time, and most importantly the
availability of 2.5D datasets allow. However,
as is characteristic of the MicroImages TNT
products, working with new 2.5D applications as they are introduced will not
require that you assemble the needed software from several pieces that are
available only on a specific platform, OS, or use one graphics engine, such as
DirectX or OpenGL. As a result you
will not become entangled in a variety of prices, release cycles, data formats,
user interfaces, and so on. All those details end up controlling the ease of use
and integration of your 2.5D geospatial applications. This may mean that other
specialized products reach selected 2.5D goals faster in some cases, but a truly
professional 2.5D geospatial analysis system requires careful integration.
Manifold Surfaces.
Definitions.
The word
“manifold” has numerous and quite different meanings in English and in
the Random House Unabridged Dictionary, for example, a manifold is a part of an
automobile engine or a pipe network. Alas,
manifold is the word already accepted
elsewhere in earth resources to refer to a curved surface in space. It
is this dictionary’s 12th definition that is most appropriate: “a
topological space that is connected and locally Euclidean.”
Note that this definition says nothing about how the manifold surface
will appear in space.
This term can be extended to cover its
use in the TNT products.
A TNT
manifold object is a raster, vector, shape, or CAD object (internal or linked)
that has been georeferenced in 3D to define its functional surface in 3D space.
Furthermore, its content must be appropriate to be viewed as a texture
projected onto the TIN or manifold surface defined by these georeference points.
From this basic definition even
more new terms have to be introduced and defined so that this new capability of
the TNT products can be
unambiguously explained here, and in future discussions of these types of TNT
geodata and shapes.
Manifold Object.
A manifold
object is a raster, vector, CAD, or shape object that has 3D
georeferencing. When a manifold
object is selected in a 3D view, including in the Spatial Data Editor, a TIN
will be formed from its 3D georeference points and the contents of the
associated raster, vector, shape, or CAD object projected onto that surface as
its texture. Georeferencing is now
both a 2D and a 3D process and is used to enter and edit the 3D control points
needed to define a manifold object. The
GeoToolbox in Display and the Spatial Data Editor can directly create a vertical
cross section manifold object. Since
a manifold object is georeferenced in 3D, it can be represented and visualized
anywhere in a 3D space, and it may or may not intersect a terrain surface.
The surface of the manifold object can be planar, curved, or creased and
folded.
Obviously the raster, vector, shape, and
CAD objects can be combined or merged in various ways before the 3D
georeferencing is added to them to define a single manifold object.
However, they can also be kept as separate manifold objects and projected
and viewed in order onto the same manifold surface.
For this purpose the 3D georeferencing of a manifold object can be
conveniently transferred from one raster object to another or between vector
objects as long as they have matching extents, just as with the georeferencing
of 2D objects. For example, you
create a raster manifold object, a matching vector overlay manifold object to
delimit boundaries, and a CAD manifold object containing annotations.
Manifold Surface.
A manifold
surface is the TIN surface formed in 3D space each time the manifold
object is used in a 3D view. The
shape of this TIN remains the same for every new viewpoint of it.
Its shape can only be changed in the Georeference process using the new
tools provided for this purpose.
Manifold Texture Object.
A manifold
texture object, or a manifold
texture, is the special name that refers to the raster, vector, shape, or
CAD object with the 3D georeferencing needed to use it as a manifold object.
The texture object is projected (called draping) onto the manifold’s
TIN surface every time the 3D view is created, refreshed, or the viewpoint is
moved.
Manifold Layer.
A manifold
layer refers to the current view of a manifold texture object draped
onto the manifold surface in a 3D view. The
3D view can have other manifold and topographic layers in it.
Terrain Surface.
You are already familiar with the use of
a terrain
surface, which is the TIN formed in a 3D view using one or more
elevation rasters. Depending upon
which of the 3D models is used, the shape of this TIN can change slightly to
refine it for each new viewpoint.
Topographic Texture Object.
You are already familiar with the use of
a topographic
texture object, or a topographic
texture, which is the 2D raster, vector, shape, or CAD object you select for
projection (called draping) onto the terrain surface every time the 3D view is
created, refreshed, or the viewpoint is moved.
Topographic Layer.
A topographic
layer refers to the current view of a topographic texture object draped
onto a terrain surface in a 3D view. The
3D view can have other manifold and topographic layers in it.
Sample Applications.
Most software products used to build and
view manifold layers and other volumes from geodata and visualize them in 3D are
highly specialized and expensive. A
popular current application is to use them to build the walls in a
seismic/geology immersion room used in oil exploration and costing hundreds of
thousands even up to a million dollars. A
picture taken inside one of these rooms can be found in the article entitled The
End of Cheap Oil by Tim Appenzeller in the National Geographic Magazine of
June 2004. In these specially built
facilities, something like a very early holodeck, experts manipulate seismic
profiles projected onto all the walls of the room as manifold layers.
Now in RV7.0 you and your
standard TNT products can begin to
explore the possibility of using these kinds of 2.5D manifold layers in your
applications.
The uses of manifold layers range from
simple graphical representations in publications to complex 3D visualizations.
Their use is common in geology; geophysics; mineral and petroleum
exploration; mining; archaeology; monitoring groundwater pollution and other
subsurface environmental properties; geotechnical, road building, and excavation
engineering; and others. To
introduce you to the appearance of these layers in a 3D TNT
view, several different kinds of manifold and associated topographic layers are
illustrated in the accompanying color plate entitled Manifolds in 3D Views.
Most of the geodata available to
MicroImages and suitable for constructing manifold objects is geological in
nature. Crisscrossing or “egg
crate” like intersecting subsurface profile manifold layers are common in
geology and are called fence diagrams. Geologic
cross sections are available on printed geologic maps and in their digital
equivalent. Any cross section can
be created from a geologic map by a trained geologist.
Seismic data is widely used but quite proprietary in nature.
As a result, the orientation of the rest of the color plates used to
describe this manifold process in this MEMO use these kinds of data sets
starting with the general illustrations in the accompanying color plate entitled
Visualize 3D Geology Using Manifolds.
Another accompanying color plate entitled 3D Subsurface Model Using
Manifolds illustrates the distribution of an ore body determined from
drilling, mine shafts, and expert interpretation.
It is hoped that you can make the extrapolation from these examples to
creating and using manifold objects from your geodata.
3D Visualization.
A raster, vector, shape, or CAD object
with the necessary 3D georeferencing is a manifold object and can be selected
for display in 3D as a manifold layer. Since
this uses the same TNT 3D
visualization models as rendering topographic layers, multiple manifold and
topographic layers can be rendered in the same 3D or stereo view. This
is a unique way of visualizing the relationships between topographic layers or
other horizontal reference layers in proper geospatial relationship to manifold
layers representing subsurface or above-surface profiles, transects, and layers
with other content and orientations.
Depending upon the viewpoint, adding
several topographic and manifold layers means that they can obscure each other.
The ordering of their 3D rendering is back (farthest away) to front
(nearest). Transparency can be used
to help you see the underlying layers in a rendering, but can create confusion
over which layer is actually being viewed.
Faster reorientation of a composite 3D, better means of temporarily
unmasking obscured layers, adding the use of manifold surfaces in TNTsim3D,
and stereo viewing are all areas of future development needed to provide more
means of visualizing the 3D relationships manifested in these potentially
complex multilayer views.
As you begin to more fully understand
this new capability, you will also find that even the 2.5D geodata needed for
this application is not widely available. However,
it can be put together from various sources starting directly within the TNT
products (for example, your geologic interpretations of 2D geodata such as bore
hole profiles and geologic maps). Unlike
all the 2D geodata already available or that you assemble, subsurface,
oceanographic, atmospheric, and other such specialized data can be hard to find,
often incomplete, and recorded in unusual units (for example, seismic
propagation times). Getting these
external 2.5D sets organized into TNT
manifold objects will require effort and potentially the use of the TNT
geospatial scripting language (SML)
for data conversion, formation, and importing into a manifold texture object
with its 3D georeferencing.
Create a Manifold Cross Section.
You can use a new tool option in the
GeoToolbox in a 2D view to quickly and easily create a simple vertical manifold
object. This is illustrated in the
accompanying color plate entitled Create Cross-Section Manifold Objects.
This procedure uses a 2D vector object overlaying an elevation raster
object in a 2D view. You use the GeoToolbox to draw any kind of connected line
representing the desired cross section on this view in 2D.
The line does not have to be straight, but must be continuous.
When you are finished drawing and editing the line, use the Generate
Cross Section icon in the GeoToolbox window to create your new manifold object.
This manifold texture object you are
creating hangs down or projects vertically upward from the trace of the
cross-section line on the terrain surface.
Thus you are prompted to enter the base level elevation of this vertical
cross-section manifold texture object in the Cross Section Options window.
This elevation value is measured relative to the base elevation of the
elevation raster object, typically mean sea level.
The value you enter will be used to position and create the edge of the
manifold texture opposite the topographic profile formed by the trace of the
cross-section line on the terrain surface.
Since you can enter any value for this elevation that is greater or less
than that of the values in the elevation raster object, this manifold object can
go up, down, or intersect the terrain surface.
The polygon boundaries in this new
manifold texture object are the map polygon boundary lines that the
cross-section line intersects in the 2D view, projected vertically down or up to
the horizontal edge you defined by the base-level elevation.
Wherever the cross section line intersects a 2D polygon edge, a vertical
polygon boundary is created. Thus,
all the polygons in this manifold’s texture object have vertical edges that
are parallel in 3D and intersect the cross-section edge at the same points as
the 2D polygons. These polygons in
the manifold texture object are all automatically assigned the same styles and
other attributes as their corresponding 2D polygons as they must match along
their intersection of the 2D vector object and the manifold texture object.
The vector object created as the manifold texture object has polygonal
topology.
The georeference control points created
automatically for this manifold object define the TIN used to shape the manifold
surface upon which the vector texture object is draped.
These control points define a vertical surface that starts at the terrain
surface. One set of points traces
out a cross-section edge that conforms to the terrain surface.
These points vary in XYZ. A
corresponding set of paired points traces out the opposite edge as the vertical
projection of these terrain surface points up or down to the plane of the
elevation you entered. These points
all lie in the same X-Y plane and, thus, have the same Z value.
Since this manifold surface is vertical everywhere, no additional
georeference points are needed to define its position in space, just these
matching pairs of top and bottom points.
Editing a Cross Section Manifold.
The simple polygon shapes in the initial
cross-section manifold texture object can be edited and shaped to meet the
below/above ground conditions on this vertical manifold surface.
Load this manifold object in the Spatial Data Editor; it automatically
displays in a 2D view and as a manifold layer for reference in a 3D view.
It will be stretched out and flattened in the 2D view by using the z
coordinate on the y axis of the 2D view. Proceed
as usual to use the Spatial Data Editor to create the revised representation of
the below ground polygons. This
editing to create a representation of this manifold texture object with the
subsurface volume features is illustrated in the accompanying color plate
entitled Edit Manifold Objects.
Any time you refresh your 2D edit view,
you see how your current changes appear when projected as a texture onto the
manifold surface in the 3D view. As
usual you can add any other manifold or topographic layers into this same 3D
view for reference purposes. For
example, if you have already designed a manifold object that would intersect
this new manifold object, you would want to view both layers in the 3D view.
If you are building a fence diagram, all polygons must match at the
intersections of the manifold surface as illustrated in several color plates.
If you have a stereo viewing device set up on a second monitor, your 3D
manifold layers and corresponding topographic layers can be rotated and viewed
in stereo as you edit their polygon shapes in the 2D view.
This cross section creation and editing
procedure duplicates the way in which most vertical cross sections are created
by drawing them. It quickly
provides an initial vector manifold texture object and its vertically-oriented
manifold surface for your use as a starting point. It creates a basic manifold
object as a starting canvas for your more complex 3D interpretations of a
geologic, archaeological, geotechnical, or any other manifold layer.
For example, this replicates the way geologic sections are inferred and
sketched and resketched by hand. But
instead of sketching them by hand, they are created as vector objects.
This provides that opportunity to view these scenarios and intermediate
results in 3D as they are created. This
is especially important when multiple manifold layers are being created that
will be interrelated in 3D.
Georeferencing a Manifold Surface.
Creating a complex manifold surface for a
manifold texture uses a collection of XYZ points in a geospatial or a Cartesian
coordinate reference system to define its surface.
These XYZ control points must be obtained by some method.
For example, they might represent the coordinates measured to various
layers in a bore hole, from towed sensor arrays, or computed from transects.
These 3D control points become the vertices of the triangles that define
the shape of a TIN onto which the manifold texture object is draped when viewed
in 3D. Using triangles to define a
manifold surface could render a faceted surface.
However, modeling a surface in this manner is very common and used in
many games, 3D viewers, and other 3D computer simulation products.
Whether or not the surface appears curved, folded, faceted, and so on
depends upon factors such as the density of these points, the designation of
edges as fold lines or some other special edge, distance to the triangle,
surface texture projection and shading technique, and so on.
In RV7.0 these 3D
georeference points and special triangle edge effects they define are added to a
raster, vector, and/or CAD object to create a manifold object using a new
procedure in the georeference process, which has been significantly expanded for
this purpose.
Choosing the new 3D Piecewise Affine
model from the menu in the Georeference window permits you to enter XYZ
coordinates for any position you select on the raster, vector, shape, or CAD
object in the 2D view. For example,
suppose the raster object represents a single, flat vertical seismic slice. To
define this vertical plane, you will only need to select the position of 4
points near the corners of the raster and enter their XYZ coordinates in the
control point table in the Georeference window. Since
these 4 points lie on a plane, the manifold surface will be defined by a TIN
with 2 coplanar triangles. If the 4
points are not coplanar, the TIN will have 2 triangles that are not coplanar and
the common edge will be a fold edge in the manifold surface.
If you want the top of the manifold surface to conform to the terrain,
then more control points are needed along the top.
If you want to curve the manifold, then you begin to enter XY values for
points that do not define a plane.
The accompanying color plate entitled Georeferencing
Manifold Surfaces illustrates the positioning of these 3D georeference
points and their tabular entries. In
this example, as is usually the case, the positions of known XYZ coordinates on
the manifold texture object are not at the corners of the object.
If no georeference points are provided at the corners and along the
edges, the TIN created would be smaller than the area of the manifold texture
object. The edge areas of the
manifold texture object would have no manifold surface to project onto and would
simply be clipped off and lost in a 3D view of this manifold object.
To preserve all the surface area
represented in a rectangular manifold object you may want to extrapolate from
the measured and known georeference points to add additional control points at
the corners of the manifold texture object.
This can be done automatically using a built-in procedure.
It extends the plane of the edge triangles defined by your georeference
points. It uses these planes to
interpolate coordinates for the corners of the manifold texture object and adds
them to the control point list. This
is also illustrated in the accompanying color plate entitled Georeferencing
Manifold Surfaces. Adding new
control points adds new triangles around the edge of the TIN to increase its
area. However, these triangles are
coplanar extensions of the edge triangles created by the 3D georeference points
you entered. You can edit the XYZ
coordinates of these added control points in the table in the georeference
window to modify the orientation of these added edge triangles.
Once you have entered four 3D
georeference points for your manifold texture object, as near to the corners as
possible, you can now open a 3D view in the Georeference process.
It will show you how your manifold surface or object is shaped and
positioned in space from any viewpoint using a wireframe view of the current TIN
or with the manifold texture object projected onto it.
Since this is a standard 3D view, any other manifold and topographic
layers can be added to it for reference purposes.
Views with multiple manifold layers are common and are illustrated on the
color plates. You can then proceed
using the 3D view as reference to edit the positions of the four 3D georeference
points or add more, add 3D interior points to shape the manifold surface,
extrapolate from the interior points to the corners, add exterior points to
expand the surface, and edit the edges of the TIN using procedures discussed in
the following section.
Editing a Manifold Surface.
Sparse Known Control.
The number of 3D georeference points you
have to enter to define the manifold surface can be few in number; a minimum of
4 are required. However,
georeference points that are known in 3D position and can be located on the
corresponding manifold texture object (the raster, vector, and/or CAD objects)
are difficult and expensive to obtain for manifold objects extending above and
below the earth’s terrain surface. The
new 3D Georeference process permits you to use your understanding of the shape
of the manifold surface represented by the manifold texture object to add in
additional XYZ control points. These
points can be used to modify the shape and edges of the TIN that will define the
manifold surface.
Expanding the Control.
You can insert control points into
the list of georeference points for the manifold surface without knowing their
XYZ coordinates or locating their positions on the manifold texture layer. In
positioning and editing these control points to improve the shape the surface
you can use anything you have available:
-
the
current 3D view of the manifold layer and various topographic layers, such as a
drape of a surface geologic map, or natural color or enhanced image overlay,
such as image ratios;
-
boreholes,
radiosond or other atmospheric observations, towed sensor arrays;
-
all
other available non-digital map, image, historical data, and lore about the area
and its volumetric relationships; and
-
most importantly, your personal expertise and skills in 3D inference.
This new feature in the Georeference
process also permits you to edit the triangles of the TIN formed from these
control points to extend and shape the outer edge of the TIN to conform to the
rectangular or irregular edge of the contents of the manifold texture object.
Matching Texture and Surface Objects.
The manifold texture object will be
larger than the coordinate area covered by the TIN connecting the 3D control
points unless there are points at each corner of the object. Conversely you may
have a manifold texture object that has null area representing irregular edges
and holes. Holes present no problem
as they automatically become transparent holes even though the manifold surface
is defined for them. Irregular or
null edges can create irregularities when this manifold layer is viewed with a
topographic layer and other manifold layers. You
may want to use a single manifold texture object that does not curve around in
space but has definite folds. You
may want to add 3D points to your TIN to better shape it in 3D space and in the
view. These special considerations
and conditions require that you edit the TIN you are creating in the
georeference process. You edit the
TIN defining the manifold surface in a 2D view.
However you can monitor the results by opening a 3D view of the manifold
object in the Georeference process. If
you have a stereo viewing device set up on a second monitor, your 3D manifold
layer and its corresponding terrain surface can be rotated and viewed in stereo
as you edit the shape of the manifold surface in the 2D view.
Available Tools.
Choosing the Georeference process opens
the table for entering 3D coordinates and a 2D view of the current TIN and
manifold texture object in flattened 2D form in the Object Georeferencing
window. Choosing the 3D Piecewise
Affine model opens a new Piecewise Control window to provide tools to specify
the special connectivity properties to use when the lines are generated in the
TIN to connect all the control points. This
new window and its use and the effects of using these edit tools are all
illustrated in the accompanying color plate entitled Editing Manifold Surface
Triangulation.
Adding Hard Edges.
You can use this tool to select any edge
in the TIN and designate the connection between these two control points is to
be a hard, or fold, edge. This hard
edge will be preserved through any activities applied to the TIN built from
these control points. Hard edges
are drawn in the TIN in a different color to differentiate them from soft edges.
Removing Hard Edges.
This tool is used to select a previously
designated hard edge and remove this special connectivity property from the pair
of control points. Thus, the
designation of this edge as hard is deleted and the points may be connected by a
soft edge in the TIN.
Remove Edges.
This tool selects an exterior
triangle’s exterior edge, thus designating that 2 control points at its end
should never be connected in the TIN. This
removes the area of this edge triangle from the TIN usually because it is null
and not covered in the irregular area of the manifold texture object or because
it obscures some other layers when viewed with this manifold layer.
Add Control Point.
This tool inserts a control point into or
outside a triangle without requiring that its coordinates be manually entered.
If the position of the point is inside any triangle, its XYZ values are
interpolated from the surface of that triangle.
It then becomes a valid control point and all the surrounding points
connected to it with soft edges. If
the point is selected outside the area of any current triangle, its XYZ position
is extrapolated by extending the plane of the nearest triangle.
New triangles are formed to redefine the edge.
In either case, once these points that have been inserted into the list
and TIN by this tool, or any other, their XYZ values can be manually edited in
this list in the Georeference window to move them from their initial
interpolated/extrapolated positions. In
this manner, the added interpolated control points can be used to shape the
interior of the TIN or the extrapolated edges.
When used together with the Remove Edges tool, you can shape the outer
edge of the TIN to match the irregular edge of the manifold texture object or
match a terrain layer.
Remove Triangles.
When a triangle is selected that has an
outer edge not shared with any other triangle in the TIN, the outer edge is
removed and its area is no longer included in the TIN.
While this tool has the same result as the Remove Edges tool, it is
sometimes easier in a complex TIN to select a triangle using its interior rather
than directly selecting its outer edge.
Extrapolate Boundary Control
Points.
This tool insures that the TIN is a
parallelogram onto which all the rectangular manifold texture layer can be
projected. This tool has been
discussed above in the section on Georeferencing Manifold Surfaces as it
is often used immediately in connection with entering the known 3D georeference
points and viewing the initial manifold surface in a 3D view. Its use is
illustrated in the accompanying color plate entitled Georeferencing Manifold
Surfaces.
Recover Deleted Elements.
This icon will undo all previous control
point editing back to the last save of the georeference object.
Using this tool with periodic saves will untangle your editing results
when the triangulation gets too complicated.
As previously discussed above in more
detail, all this edit activity with these tools is performed on the current
representation of the TIN formed from all the current control points and their
connectivity relationships. However,
while you appear to be adding and removing triangle edges, you are actually
merely establishing new relationships between pairs of points in the 3D
georeference subobject for the manifold texture object(s).
Create a Manifold Object Using a Script.
Both a manifold texture object and its
manifold georeference subobject can be created using new classes and functions
added for this purpose to the TNT
geospatial scripting language (SML).
Using these tools, you can build a complete manifold object in geospace
or in some arbitrary Cartesian coordinate space.
The manifold texture object can be imported or computed from equations
using SML.
Similarly the control points to define a manifold shape can be imported
or computed from equations The
simplest application would be to convert a large tabular collection of XYZ
points, too numerous to enter by hand, into a large georeference table. The
introduction of these new classes and functions along with more suggestions for
their uses occur in the section below entitled Geospatial Scripting Language
(SML).
Backward Incompatibilities.
If any object is georeferenced with 3D
points only in RV7.0 for use as a
manifold object, that object and the Project File that contains it can not be
used in any V6.9 or earlier TNT
product. The reasons for this are
identical to those explained above in detail in the section entitled Coordinate
Reference Systems / Backward Incompatibilities.
To avoid this limitation, it is suggested that any objects you intend to
georeference in 3D for use as manifold objects be kept in a separate Project
File if you are also still using V6.9
of your TNT product.
Cartoscripts.
MicroImages’ Reseller in Italy has made their collection of geologic line styling Cartoscripts available for
you to download and use at www.spaziogis.it/index.php?
option=com_docman&Itemid=71. While
these represent lines accepted for use in Italian geologic maps, you can use
them as starting points and modify the scripts for your own nation’s map
conventions.
Georeferencing.
There have been major changes in this
process and these are discussed in detail in the section above entitled Manifolds.
Some of these changes, such as the ability to name your control points,
will effect you if if you are not using any manifold operations in this process.
Raster Resampling
Using Georeference.
This
interface has been redesigned to use a window with tabbed panels.
The Rasters tabbed panel shows the list of selected rasters along with
properties, including dimensions, georeference extent, and Coordinate Reference
System. The Settings tabbed panel
shows all available settings so there is no longer a need to browse each of the
menus to see what the various options are set to.
When resampling to match a reference
raster in cell size and orientation, the output raster cells will now exactly
align with the reference raster cells regardless of the Extents setting.
This is useful when there is only a partial overlap with the designated
reference raster.
When resampling to geographic
(latitude-longitude) coordinates, the cell size is now specified in
degrees/minutes/seconds. As usual,
values may be entered in decimal degrees, though they will be shown for the
cells in degrees/minutes/seconds.
Options are now available for choosing
the pyramid computation method. Available
choices are None, Average, Sample, and Automatic.
The default Automatic choice will use sampling when the resample method
is nearest-neighbor and averaging when bilinear or cubic convolution is used.
The reference raster and related settings
are automatically reset when a new set of input rasters is chosen.
A null mask subobject is no longer
created for the output raster if the input raster does not have a null value
defined and set.
Raster Mosaic.
The
TNT products are superior to any
other in displaying and managing massive images.
Integrating all the pieces of your images (for example, DOQQs or
orthoimages) into a single, large layer can be very convenient for your further
analysis, for use in a TNTatlas, TNTsim3D,
TNTserver, and so on.
This is even the best procedure if your ultimate objective is to break up
the image into new, smaller geographic units after your analysis.
A common example is to assemble orthoimages into a single, large image
with a common coordinate system, work with this mosaic, and then divide it into
new pieces and the map projection required in some other system.
Your use of the mosaic process to create
these massive but convenient images has been discussed briefly in the Editorial
section above. Additional detailed
information about the creation of compressed massive images was discussed in
detail in the MEMO entitled Release of RV6.9 of the TNT products and
dated 31 December 2003
in the section entitled JPEG2000 Compression.
RV7.0 of the mosaic process
has no dramatic change in its interface or many features.
However, considerable attention has been addressed to increasing its
robustness and optimizing its performance for these large tasks.
Null Mask.
Mosaic now automatically generates a null
mask layer and its pyramids to define non-data areas and null cells in the
output raster. It, and any other
process that displays this output raster, will automatically use this null mask
as will other processes converted to do so.
This also significantly improves the mosaic’s handling of input rasters
that have no null value or different null values.
First Raster Overlap.
An overlap mode of First Raster has been
added whereby cells from the first raster in the list covering each output cell
will be used. The mode of Last
Raster was available in RV6.9 and
earlier, but now you do not need to reorder your selections to use the first
raster.
* Output to JPEG2000.
RV6.9
of the TNT products required that
you mosaic your raster and then use the Extract process to apply JPEG2000
compression to it. The mosaic
process in RV7.0 permits you to
apply lossless or lossy JPEG2000 compression to your mosaicked output raster
object. The many objects you select
for input to mosaic can be any of a mixture of uncompressed or objects
compressed with JPEG, JPEG2000, or in any mixed coordinate reference system
(CRS). Note that DPCM, or
Differential Pulse Code Modulation, is now called Standard Lossless compression
in TNT dialogs.
Also selected and mixed in can be linked external rasters with or without
compression such as GeoTIFFs, MrSID, ECW, TIFF, and others as long as they are
georeferenced and their CRS identified internally or via an auxiliary file.
To permit this flexibility, the Mosaic process still requires that it be
provided with adequate drive space to create the entire mosaicked raster object
in its final uncompressed and also compressed form if compression is specified.
Also, as discussed earlier in this MEMO, using tiles during the
compression procedure to avoid creating temporary storage for the full
uncompressed raster then compressing it, can have detrimental effects at the
tile edges and legal ramifications. The
accompanying color plate entitled Mosaic Directly into JPEG2000 provides
sample results from mosaicking 100 GeoTIFF images directly into a single
JPEG2000 raster object.
Temporary Storage Requirements.
In RV7.0
each input raster object is uncompressed as needed and resampled into this large
raster object using your desired CRS, sampling method, overlap method, and so
on. The size of this uncompressed
temporary mosaic can be closely estimated in advance from the extents of your
input objects, compressed or not, and your setting for the mosaicking process.
However, if you are requesting that your mosaic produce a compressed
output raster object (for example, a lossless JPEG2000 result), this temporary
raster must coexist in your storage system at the same time as the uncompressed
result. However, the exact size of
a compressed raster may not be accurately estimated.
It can be closely estimated for a JPEG2000 compression (for example a
request of 10:1 lossy compression). However, the size of a mosaic object using
JPEG compression or JPEG2000 Best Quality is not so easy (for example, a request
for a 75% JPEG lossy result). If
you target a compressed raster output, then at the onset the mosaic process will
liberally estimate this additional storage requirement and warn you if the
temporary and total drive space available would likely be insufficient to
complete the final compressed or uncompressed mosaicked object.
Sample Applications.
The following kinds of test applications
have been used to challenge and confirm the robustness of the Mosaic process
especially for producing a JPEG2000 compressed raster object.
Lossy JPEGs into Lossless JPEG2000.
Lossy compression should never be applied
to images to save storage space when those images will be used for some form of
computer image analysis. However,
using lossless JPEG2000 or standard lossless compression (formerly called DPCM)
can be specified for a mosaic result intended for additional analysis. Keep
in mind that Mosaicking images, depending upon your settings, may alter them by
resampling, by feathering the edges, and so on.
However, many orthoimages, especially those collected by digital cameras
from the air, are being distributed as JPEGs or as lossy compressed TIFFs.
To avoid excessive additional changes in these already lossy compressed
images, you might want to link to and select them as input for your mosaic and
choose lossless JPEG2000 as the compression method for your mosaicked
raster object.
Over 6000 units of color Digital Ortho
Quarter Quads (DOQQs of 3.75' by 3.75' ground areas) are available from the site
of the Nebraska Department of Natural Resources.
http://www.dnr.state.ne.us/databank/fsa03.html.
These DOQQs were acquired using an airborne digital imaging system and
are in 24-bit color with 1-meter resolution.
They are provided for downloading in both UTM and Nebraska State Plane
CRSs using JPEG compression. Remember,
JPEG is always lossy compression! Although
the JPEG compression is about 20 to 1, no JPEG artifacts are evident and image
quality color balance between units is good.
These JPEG DOQQs (*.jpg) files are each provided with a georeference via
a companion world file (*.jgw) file of the same name.
It is a simple matter to select hundreds of them in mosaic, which will
autolink to them. Mosaic can then
assemble them into a single JPEG2000 lossless output raster object representing
a county or some other project unit. Since
the input objects were heavily compressed JPEG files, the lossless JPEG2000
mosaic is also of a significantly reduced size.
A future development in mosaic would be to permit it to use a region
defining the irregular boundary of a county or project to control the portion of
the input objects to mosaic and even to automatically select the objects needed
from a common directory(s).
Lossy MrSID into Lossy JPEG2000.
Landsat 24-bit color orthoimages of
15-meter resolution can be downloaded for almost all the earth except for the
poles for circa 1990 and circa 2000. These
NASA sponsored orthoimages are in 6 by 6 degree units and are compressed about
30 to 1 into MrSID files. These
MrSID images (*.sid) can be downloaded with their georeference provided in a
companion world file (*.sdw). Mosaic
can then link to these files and convert them into a single province- or
country-covering raster object and apply lossy or lossless JPEG2000 compression.
Now you can prepare and substitute a 15-meter color image coverage of
your nation in place of the 1-kilometer MODIS images distributed with RV6.9
on the Global Reference Geodata DVD. Using
reasonable compression for your JPEG2000 results will permit you to use these
large coverage images in your projects including their free distribution with TNTatlas,
TNTsim3D, and via TNTserver.
Predefined Raster
Combinations.
Statistics for corresponding cells in a
set of rasters can be computed using the Statistics operation under the
Algebraic combination category. This
procedure computes two output raster objects, one containing the cell-by-cell
mean and the other the cell-by-cell standard deviation for the selected input
raster objects. This operation
might be used, for example, to compute cell-by-cell statistics for a time-series
of vegetation index rasters (or rasters containing any other measured or
computed numerical value).
An Exclusive Union (XOR) logical
operation can be performed on a set of binary raster objects.
Like the previously-available logical AND and logical OR operations, the
XOR operation produces a binary raster object.
When performed on a pair of input rasters, XOR produces a cell value of 1
if either input has value 1, but not both.
Thus the result is 1 if the two corresponding input cells have different
values (0 and 1) but 0 if the two input cells have the same value (1 and 1 or 0
and 0). When you select more than
two input objects, the operation is performed sequentially: input 1 XOR input 2,
result 1 XOR input 3, result 2 XOR input 4, and so on.
In general, the cell result will be 1 where the number of input 1's is
odd and 0 where the number of input 1's is even.
Raster to Vector
Boundary.
This process will now warn you if you are
requesting it to convert a highly variable input raster into a vector object.
Since this is a somewhat qualitative appraisal by the program, you can
override this warning and attempt to proceed.
Converting a raster object to a vector
object is generally applied to categorical or theme rasters or to the results of
automated image processing which categorizes the image into a few classes
grouped into clusters of cells of equal value.
It is not usually meaningful or useful to convert a highly variable
raster object such as an unprocessed image to a vector object.
Representing a cell as a vector polygon with topology will inflate the
storage for that cell by about 100 times. This
is not due to the vertices for the polygon but the overhead associated with the
definition of a polygon. This
overhead is not increased appreciably when the polygon bounds a meaningful area
of uniform cells, but is prohibitive when each cell or a small number of cells
make up each polygon. It is also
very time-consuming to make this large number of conversions.
Thus, this inappropriate kind of conversion will either take so much time
to complete that you would conclude that it was not working or produce a huge
object. Another result would be
that it would run out of storage space since the storage can not be estimated
before the vector object is complete—it could easily grow to 100 times the
count of the cells in an image.
Import/Export.
Vector Import.
CARIS ASCII graphical data files can now
imported. CARIS is a vendor of
GIS-oriented products focused upon hydrographic and coastal applications.
MapInfo MIF files now import with
improved support for text fonts, color, and styles.
Vector Export.
Geography Markup Language (GML) files
conforming to version 2.x can be exported from vector or CAD objects.
GML (*.gml) is the OpenGIS’s XML encoding for the transport and storage
of geographic information, including both the geometry and properties of
geographic features. Specifications
for GML can be found at www.opengis.org/ specs/?page=specs.
ESRI shapefile export can now export
vector or CAD label elements as shapefile point elements with the label string
as the database record entry for the point.
Raster Import / Linking.
ESRI ArcGrid files can now be imported in
all forms including even when they are compressed.
Images in the Robinson projection can now
be imported from the ArcGrid format.
PNG (*.png) import will now recognize the
ICM color profile tag in the PNG file and import the ICM color profile as a
subobject of the raster object. If
the ICM color profile does not exist but the gamma and chromaticity values do,
then an ICM color profile object is created under the raster to contain the
gamma and chromaticity values.
MrSID files (*.sid) can now be imported
or autolinked on Mac OS X, Windows, and Linux/UNIX.
In V6.9 this capability could be used
only on Windows-based platforms.
TIFF import and auto-link
will now convert the metadata in the TIFF directory to a metadata
subobject in the Project File.
Sun Raster (*.ras) import will look for
the companion world file (*.snw) of the same name and transfer its georeference
into the corresponding raster object.
Raster Export.
When export georeference to
“descriptive text” all Coordinate Reference System (CRS) details are
included.
PNG
export process now exports the opacity mask for the file as the alpha channel.
The export will also export the ICM color profile if one exists under the
raster object.
TIFF export now optionally provides
“pack bits” compression for 8-bit rasters and uses it by default
instead of LZW compression.
GeoTIFF export now optionally creates a
conamed companion world file (*.tfw) containing its georeference data.
Some other products read a GeoTIFF as a TIFF and expect the georeference
information to be available in this world file.
The scale factor for the Lambert Conformal Conic projection is now also
included in the GeoTIFF.
Font Management.
Previous issues of this MEMO have
discussed in detail the issues of font management that plague this industry due
to font copyrights, varied font rendering methods, and other similar issues some
of which must date all the way back to Gutenberg.
Your problems arise when you move projects and results between various
computers and are even more likely when the move is between platforms with
different operating systems. This
even occurs when your cross platform activities (for example, Windows to Mac)
are totally within the TNT products.
The new location or operating system may not have the same fonts
available and you may have used one that can not be moved due to a copyright,
can not be rendered well on the device (paper versus monitor), and so on.
At that point the software at the destination has no choice but to make
font substitutions with or without your input.
Font substitution is likely to occur in
published results such as the TNTatlas,
SVG layouts, and PDF documents that can be designed to run across all platforms.
To minimize the impact of this, MicroImages has continued to improve how
font substitutions are made within the TNT
products and your project results. Two
forms of font substitution are now used in RV7.0:
“considerate automated substitution” or in the order of your preprogrammed
list of “designer specified substitution.”
Considerate Automatic Substitution.
TNT
font substitution is automatically used when you have specified only one font
and it is not available. It is
“considerate” because it does not make arbitrary substitution, which is
common, but matches up fonts that have similar appearance, for example sans
serif fonts are replaced by other public domain sans serif fonts.
These TNT substitutions are
preprogrammed for various font styles. The
substitutions are listed in a table on the accompanying color plate entitled Font
Substitution in the TNT Products.
Designer Font Substitution.
When you choose a font name or family
during the design of your project, you are attaching that font name to that text
string (for example, {~FArial-Bold.ttf}), which would render the character
string with this format code in Arial Bold.
However, you can define your preferred substitution using the form
{~Farialbd.ttf,Helvetica.dfont;2,Arial-Bold.ttf}.
This will be interpreted by TNT
processes to mean use the Microsoft version of Arial bold if it is available, if
not try for Helvetica, then try for the Sun version of Arial bold.
If none of these fonts can be found, then automated substitutions will be
made. The superior results of using
designer or considerate automatic font substitution for a DataTip created with
Arial fonts in Windows and then moved to a Mac to find and use Helvetica is
illustrated in the accompanying color plate entitled Font Substitution in the
TNT Products.
Map Calculator.
The map calculator use and operation
remains about the same as RV6.9.
However, it now takes complete advantage of the new Coordinate Reference
System (CRS) introduced in RV7.0.
See the above major section entitled New Coordinate Reference System
for details on these many new, possible conversions.
* Advanced
Geometric Object Conversion.
Diverging
Objectives.
CAD
software has a very large user base; why doesn’t this software dominate the
GIS application area? Vector
topology provides the basis for many useful area oriented operations; why is it
not maintained in GIS systems? Geo-oriented
games have the largest installed user base of all and have their own graphical
storage structures. IT managers are
in charge in large, centralized enterprises that build their business
infrastructure on databases software including any geographical data.
Web servers have their own preferred graphics formats, which they try to
adapt to fit their geographic requirements.
All
of these important spatial endeavors are built up on graphical data structures.
However, all these various structures are quite different since their
primary application, while it may be spatial in nature, is quite different.
All of you have experienced this in attempting to convert CAD data to GIS
data and vice versa in some software package.
It can be done, but the results can be unusable.
Converting a CAD drawing with many blocks and individual lines with their
own part description into a vector object can produce a very bloated vector
object with complex topology. Converting
a vector object to a CAD object can also produce complex results as everything
can be in pieces and no concept of organized subparts, such as blocks, will be
represented. In fact, shape type data storage structures evolved into existence
to better fill the gulf in the middle of the CAD, vector, and database
structures.
Resulting
Complications.
Alas,
if you are truly engaged in geospatial analysis, not just isolated in GIS,
engineering, database management, … you will quickly learn the value of these
different structures. You also
quickly learn that the geospatial analyst has to deal with each and all of them
with facility. A typical simple
urban project will have property information in CAD form, ownership records in
some database structure, historical data such as deeds in paper form or as
scanned PDFs; requirements to upgrade it all with MrSID, JP2, or JPEG images;
images of buildings in JPEG; a desire to go online to serve its internal needs
(intranet) or public access (Internet); and so on.
And this simple problem is repeated over and over worldwide with varying
degrees of existing data preparation and quality.
Major nations do not even have a documented cadastre or records of where
things are on or in the ground. These
kinds of issues get much more complicated when the source materials that do
exist are not only in these different formats, which are “optimal” for their
current use, but also jealously guarded in that format for compartmentalized
business or military turf or other compartmentalized reasons.
Integrating
Disparate Strategies.
All
of this is already clear to many of you who, by buying your TNTmips,
understood the need to deal with using each of these project components in its
optimal form. You also realize that
at some point in a geospatial analysis process, you will need to move geodata
between these formats with as much facility as possible or use them together
such as in a TNTatlas.
Perhaps this is merely a requirement to link to that data structure and
add it as a display layer. Perhaps
you need to edit that data structure. Eventually
you will need to move from one structure to another.
As
an encompassing geospatial system, not a GIS, CAD, database … system, TNTmips
and its associated products are focused upon meeting all these diverse
requirements. RV7.0
makes significant gains in improving the integrated use of these common, but
marketed different data structures that you will encounter and use.
This is an evolutionary process and never will be truly easy.
Gradually TNTmips is
supporting the direct use of these data structures as layers, without import,
directly from their various native formats.
Shapefiles can now be used in this manner.
JPEG and PNG rasters have also been added to this list.
Geometric
Conversion Engine.
Improved
Integration.
RV7.0
introduces into several key processes a new Geometric Conversion Engine, which
consolidates the fragmented code used in the processes to convert between
vector, shape, CAD, TIN, and region data structures.
This immediately made it possible to create and plan for many new
features in those TNT processes that
previously only extracted and/or copied geodata within a specific object type.
Using this new engine, Copy and Paste and the processes to Extract and
Merge can now automatically convert data between these geometric data types (for
example, CAD to vector). This
powerful new engine also incorporates all the “transparency” properties you
expect in a TNT process.
For example, the Coordinate Reference System (for example, projection) of
the source data (which means, the copied area) may be different from that of the
destination (which means, the paste layer) and will be converted as it is
inserted.
Expanded
Capabilities.
In
these reincarnated processes, you can use any of the TNT
selection procedures to determine what source elements to use in the conversion
process. This inherent TNT
capability also takes advantage of the new and greatly expanded Coordinate
Reference System released in RV7.0
and discussed elsewhere in this MEMO. You
can now also elect to use a complex region boundary to define the area to be
converted in these processes in addition to how the edge effects should be
handled (which means, partially inside, completely inside, …).
And bigger objects being converted to vector form use the faster and more
powerful validation engine and its new abilities to resolve conflation problems
when the source and destination objects have many complex overlapping features
as discussed in detail elsewhere in this MEMO.
With these powerful new capabilities now in place in RV7.0,
you can expect more capabilities along these lines to appear in the TNT
products to further define what a geospatial analysis system should be capable
of. The current new conversion
capabilities are summarized in the following subsections and a sample is
illustrated in the accompanying color plate entitled Geometric Object
Conversion.
Faster
Operation.
Merge, Combine, Extract, and copy/paste
operations to a vector object is now faster due to improvements in the
topological validation step. While
this step is faster in RV7.0 its
result is greatly improved as it is checking for and resolving conflation
problems in the output vector object.
*
Geometric Object Conversions.
Using
the new integrated common code Geometric Conversion Engine “Vector to CAD,”
“CAD to Vector,” “Region to Vector,” and “TIN to Vector” processes
have been replaced by a new set of processes called “Geometric to Vector,”
“Geometric to CAD,” and “Geometric to Region.”
The term “Geometric” refers to any spatial object that is not a
raster, raster set, or hyperspectral object.
The current geometric set includes vector, CAD, TIN, shape, and region
objects. Therefore, the new
“Geometric to Vector” process will allow selection of CAD, TIN, shape, and
region objects to convert to vector objects. “Geometric
to CAD” will allow the selection of the vector, TIN, shape, and region objects
to convert to CAD objects. “Geometric
to Region” will allow the selection of vector, CAD, and shape objects to
convert to region objects. All of
the “Geometric to…” processes now support selection of multiple files.
The
Geometric Conversion Engine that is used in the conversion processes can accept
subsets of the source objects using the options “By Script,” “By
Attribute,” and “By Element” to generate which elements to convert.
“By Script” is used to select elements via query, “By Attribute”
is selecting elements via record attachment, and “By Element” uses the
elements selected via the user interface. These
all work in conjunction with the region selection capability.
For example, you can choose a CAD object, select some of its elements via
a script and limit the area using a region to convert or merge to a vector
object.
In the Geometric to CAD process if a
vector object is selected, a line is created in the destination CAD object if
the vector line has an attachment to a record in a user defined table, even if
the vector line is part of a polygon that was transferred to the CAD object as
well.
*
Merging Objects.
Process
/ CAD / Merge… and Process / Vector / Merge… now allow selection of other
geometric objects to be merged into the destination vector or CAD object.
These geometric objects include vector, CAD, TIN, shape, and region
objects.
The
Geometric Conversion Engine that is used in the Merge processes can accept
subsets of the source objects using the options “By Script,” “By
Attribute,” and “By Element” to generate which elements to convert.
“By Script” is used to select elements via query, “By Attribute”
is selecting elements via record attachment, and “By Element” uses the
elements selected via the user interface. These
all work in conjunction with the region selection capability.
For example, you can select a CAD object, select some of its elements via
a script and limit the area using a region to convert or merge to a vector
object.
The vector Merge process now allows the
selection of the topology type (polygonal, planar, and network) for the
resulting vector object or the merge process will automatically determine the
best topology level given the source objects.
* Vector and CAD
Extraction.
These Extract processes are now
integrated into the Geometric Object Conversion process discussed above.
They can still be accessed on the menu under this older entry to avoid
confusion.
CAD Object
Warping.
CAD object warping now supports
densification of the elements.
Spatial Data
Editor.
* Copy.
The Copy / Paste operation now has
significantly expanded capabilities using the new Geometric Object Conversion
engine discussed above. You can now
copy from a vector, CAD, TIN, region, or linked shape object (which means,
shapefile) including linked objects. During
the copy operation, you can select the irregular area to copy from any using a
region object. This copy area
selected from a vector, linked shape, CAD, or TIN layer can be optionally
controlled as appropriate to the source object type to be Partially Inside,
Completely Inside, Clip Inside, Partially Outside, Completely Outside, and Clip
Outside the source object. An
example of the application of this feature is illustrated in the accompanying
color plate entitled Copy/Paste between Geometric Objects.
* Paste.
Regardless of the type of geometric
object selected for the copy operation, the subarea can be pasted into a CAD
object or into a vector object and their relational database structures will be
reconciled. The topology will also
be validated if the target is a vector object.
Future capabilities will permit other geometric objects as the generic
paste destination (for example, a shape object).
This strategy is another big step forward in permitting your TNT
application to understand, interchange, integrate, and take advantage of the
unique and useful properties of each of these widely, but usually separately
used geometric data structures.
Toggle Through Mixed Elements.
Right Mouse Button operations will now
allow you to toggle through nearby elements (not just vector element types as in
previous TNTedit versions) for
vector and CAD editable objects. For
example, two or three vector labels lie on top of each other as a result of
label auto generation. Pressing the
right mouse button will select one of the labels, then pressing the tab key will
toggle through the other labels and any lines, polygons, nodes, and points that
are nearby, allowing the selection of the label you want to move.
For CAD elements, holding the right mouse button and pressing the tab key
will toggle through all of the CAD elements it found within its search distance
allowing you to edit elements that are hard to select.
These time saving actions are illustrated in the accompanying color plate
entitled Step through Elements with Tab Key.
Manifolds.
Creation and / or editing of a spatial
object with a manifold georeference is now supported.
When opening a reference or editable spatial object with a manifold
georeference, the editor will generate a set of dialogs for a separate 3D view.
This 3D view will contain a layer referring to the manifold georeferenced
object in the Editor’s View window. Any
other reference layers can be viewed in the 3D view window.
Creating a new object over the reference manifold georeference object
will copy the manifold georeference to the new object.
All reference and editable layers in the Editor’s View window must have
the equivalent manifold georeference. When
all of the layers that have a manifold georeference are removed from the edit
view, the associated dialogs are closed.
Undo.
Changes made in the Numeric Edit tool can
now be reversed using the “Undo” button.
©MicroImages, Inc. 2010 Published in the United States of America
11th Floor - Sharp Tower, 206 South 13th Street, Lincoln NE 68508-2010 USA
Business & Sales: (402)477-9554 Support: (402)477-9562 Fax: (402)477-9559
Business info@microimages.com
Support support@microimages.com
Web webmaster@microimages.com
| 