Browse Prior Art Database

Image Processing Context

IP.com Disclosure Number: IPCOM000105009D
Original Publication Date: 1993-Jun-01
Included in the Prior Art Database: 2005-Mar-19
Document File: 4 page(s) / 133K

Publishing Venue

IBM

Related People

Boyd, C: AUTHOR [+4]

Abstract

The Image Processing Context is described as a user configurable graphical image processing pipeline for rendering geographical image information. The pipeline is configurable because the user may specify the order in which the image processing operations are performed as well as the parameters describing each processing step. The pipeline may be configured to provide any subset of the graphical image processing operations provided for in the implementation of the pipeline.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 49% of the total text.

Image Processing Context

      The Image Processing Context is described as a user
configurable graphical image processing pipeline for rendering
geographical image information.  The pipeline is configurable because
the user may specify the order in which the image processing
operations are performed as well as the parameters describing each
processing step.  The pipeline may be configured to provide any
subset of the graphical image processing operations provided for in
the implementation of the pipeline.

      The pipeline operates, in the forward sense, with each
processing step producing a new output image which is then passed to
the next processing step.  Intermediate images are discarded
automatically by the pipeline.

      The pipeline is intended to be used to prepare a wide variety
of image data for display or printing.  It accommodates binary,
greyscale, and coded color imagery; it can be applied to spectral
band imagery by treating each band as a single grey scale image.

      Configurations for the pipeline may be created, modified and
discarded by the user and are maintained in a pseudo SQL table for
this purpose.  Once a configuration is created it may be re-used or
modified as many times as desired.  The configuration itself is
called an Image Processing Context.

      Since the source image data may reside in files or memory, as
specified by the user, several pipelines may be "joined" with the
output of one pipeline becoming the input to another pipeline.

      In addition to the image data itself, statistics and
information about the image are also carried through and modified by
the pipeline.  This information includes cartographic description,
position, scale, color encoding information, size and pixel data
type.

      The pipeline is not limited to geographically coded data and
may be used to process non-spatial imagery.  Of course, if the image
is not geographically encoded it cannot be cartographically
projected.

      In Fig. 1, the source image is retrieved from a file and
processed through a cartograhic projection and color encoding in the
first pipeline.  The output of the first pipeline is saved in memory
for subsequent reuse by the second pipeline.  The second pipeline has
been configured to retrieve its data from the output of the first
pipeline, performing a scaling, rotation and subimage transform for
the final output image.  The second pipeline may be quickly updated
under program control to provide the ability to implement plane
geometric operations for the display (commonly known as rotate, zoom
and pan).

      Many different configurations are possible and may be tailored
to unique requirements of the application and data.

IMPLEMENTATION - The graphical image processing operations provided
in the implementation of the model are:

o   Rotation

o   Scaling

o   Cartographic projection

o   Sub-image extraction

o   Color encoding for the look-up tab...