which is about thinking spatially...
... and being able to actually do something about it
more than was possible before:
computers, and remote sensing satellites, and GPS systems
Micha Pazner
What’s Yours?..
GeoPornography
The Geography of Pornography
Announcements
People not registered and who are interested in taking this course:
Please talk to me during the break or at the end of this lecture.
Today’s Meeting:
The Course Outline
How the course fits into the larger picture
What are we here for?
A bit about what I do
Spatial Image Processing (SIP): Data
Course Outline
(1) notes , transparencies, and blackboard can be used to talk about the course outline’s contents.
(2) Distribute the outlines to the students
read carefully and ask me questions next time you see me
GIS 2000
A unique opportunity to inject a bit of:
"learning by professional travel"
outside of the confines of our course and classroom
www.gis2000.com (Check it out!)
Benefits:
Exhibit Hall (vendors, colleges, some freebies)
Mini Conferences
Poster Presentations
Special teacher/student program
A sense of the ‘Professional Community’
In sum:
instructional value
career value
Extra Curricular activities
You have 3 options:
Go on a 2-day trip (At an estimated cost of $150.-)
Go on a 1-day trip (At an estimated cost of $75.-)
Stay in London
Those that wish to participate:
I need next week (class time)
a written note with your name and 1-or-2 day preference,
plus a $ 20.- deposit
Questions Please!:
How the course fits into the larger picture
The Techniques Stream in the Department
Other opportunities for making use of GIS knowledge in your program
Application areas
Field Camp
mine next year is the "Adirondack Mtns, New York"
This trip to the "Forever Wild" region will have the following elements:
tour the area by van,
by foot - hiking
camping, camp fire evenings
canoe,
climb a mountain,
go underground, and last but not least do some:
seaplane recon.
Honor’s Thesis
Publications
Joint publications with undergraduate, not graduate students
here are a few examples:
Pazner M. and B. Reynolds,
The "Clear Box" Image Processing Simulator: A Physical Device for Demonstrating Digital Image Processing Functions, extended abstract, published in CD-ROM, Third International Symposium for Integrating Geographic Information Systems and Environmental Modeling, Santa Fe, NM:NCGIA January 21-25 1996.
Pazner M. and P. Stephenson,
Global AIDS Relative to Population: A Map Depicting Population, Total AIDS Cases, and AIDS-per-Million for the Countries of the World, a multivariate map, London, Ontario: UWO & ThinkSpace, February 1995, tabloid sheet.
Ripley N. and Pazner M.,
Enriched Transect Profiles in a Raster GIS Environment, Poster. Canadian Cartographic Association and Association of Canadian Map Libraries and Archives (CCA/ACMLA) Joint Conference, London, ON, May 27-30, 1998.
Lafreniere M., M. Pazner and J. Mateo,
Iconizing the GIS Image, GIS/LIS '96, pp. 591-606, Denver, CO: AAG, ACSM, AM/FM International, ASPRS, URISA. November 1996.
What are we here for?
Please write a "1 minute essay" on:
What are we here for?
What are we here for?
Learning, Thinking, Learning to Think, Learning to Learn
A quote adapted from Guy Allen, 1992 3M Fellow, UofT
Learning is a sacred activity
It’s power to affect people’s lives–
positively or negatively–is overwhelming.
To be a good student, you need love, energy,
clarity, idealism and realism–
something of the spirit, and something of the earth.
A bit about what I do
Academic Background
College/University degrees in –
Computer Science (Prac. Eng., 3 years), Economics (B.A., 4 years), Geography (M.A., 3 years), Geography (Ph.D., 4.5 years)
[14.5 academic years folded into 12.5 years]
Currently I am an academician:
specializing in Geographic image processing
doing: teaching, research, admin
Professional milestones as a prof:
Accidental academic software developer:
MAP II Map Processor team
The MAP II project led to 3-4 years of work with The World Bank
mainly in Angola, Africa.
Co-author of a GIS book entitled "Simple Computer Imaging and Mapping" (Pazner, Thies and Chavez)
Co-founder of ThinkSpace Inc.
a GIS R&D corporation
housed at the UWO Research Park
products: Map•Factory and MF Works
Questions or Comments?
Introduction to Spatial Image Processing (SIP)
Announcement:
I urge some of you to take notes during the lectures...
an external representation that can help "burn" an internal one (mental notes)
This lecture unit describes the data model, and goes on to introduce the spatial image processing (SIP) operation types in their groups (next week)
Definition: Spatial image processing operations process remote sensing, cartographic, and field data organized in pixel-image-stack form.
SIP operations are commonly referred to as "raster GIS operations"
and sometimes as "map algebra".
Another name could be a "layer calculus".
Spatial Image Processing (SIP) Data
The Data Model and Structure
The data model of image-based (raster) GIS is multilayered numeric arrays
i.e.: an aligned stack of layers, where
each layer is made up of an array, and
each array consists of square cells
* Transparency: The data model *
One way to think of the stack is to imagine a:
stackable shelving unit constructed from transparent material (eg. plexiglass) that can hold aligned transparent images.
here’s a transparent shelf that represents a layer that holds an array...
and here is the ClearBox image processing simulator...
ClearBox is a physical device we created for use as an instructional aid in the classroom
to take some of the ‘black box’ flavour out of image processing..
Basic Data Assumptions, Conventions and Terms
Basic assumptions about the data are needed...
...and common conventions and terms must be layed down.
This will be done in hierarchical fashion, from low level to high level, starting with a digital Value.
[with Computer and ClearBox and Blackboard]
Values
Values are numbers usually represented in the computer as integers and sometimes as decimal (floating point) numbers.
The empty set is represented by a special null value termed Void.
The value is a component of a cell.
The digital value of a cell lends itself to powerful manipulation.
Values can be used as labels (nominal, value as noun) or as measures (numeric, value as quantity).
A given value Z is stored in a cell.
Cell
Each Cell has a value Z.
Cells are square ‘pieces of space’, much like tiles.
Cell resolution is the length of a cell side in real-world units.
The location of each cell is a certain row X and column Y in an array.
A group of one or more cells sharing the same value Z is termed a zone.
Zone
A group of one or more cells sharing the same value Z is termed a zone.
Zones can be disjoint, i.e.: with cell members scattered in the array.
Zones are commonly also referred to as ‘regions’, ‘classes’, or ‘categories’.
Zones can represent point, line, area, or surface features
By definition, all the zones in a given array fill up the entire array and do not overlap.
Taken together, the zones fill a retangular array termed a layer.
Layer
A Layer is an array of cells, each with location (X, Y) and value Z
The locations define a regular square grid.
A cell can be viewed as a tile occupying a grid square.
The grid origin is at the Upper Left, as opposed to the Lower Left cartesian coordinate system used in graphs.
The layers have a single fixed cell size or cell resolution.
The layers are at the same cell resolution, ie. the same ‘data-stucture scale’
Each layer has a name and associated with each layer, is a legend.
Legend
The legend is an index in list or table form of the zones (ie. types of cell values) found in the array layer.
Each zone entry in the legend is considered a category.
The legend establishes a correspondence between spatial zones and thematic categories.
Characteristics of a zone or category are recorded and displayed in the fields of each legend entry, such as: value, color or pattern, label or text, and cell count (frequency).
The legend is a dual data structure to the layer
The legend , with its categorical entries, is an important information construct that accompanies each array layer.
The legend is a complementary (dual) data structure to the layer, with its categories corresponding to layer zones.
Changes in one structure are reflected in its dual.
The tight coupling of map (layer) and legend is a unique cartographic legacy
that has special consequences for image processing.
For instance, users can manipulate one or more legend categories in order to achieve a desired effect in the array layer.
therefore the legend is an active data representation which plays an integral role in the processing of the information.
Stack
Layers are organized in a higher level construct–the stack.
However, the stack is unordered
Therefore, the stack is best viewed as an organizational facility used by the system to handle collections of stackable layers
Made up of a higher data structure and procedures for managing it
For instance, the "Project Manager" in Map•Factory
Basic requirements for layers to be included in a stack
The layers in a stack must share the same geometry, have the same grid reference and cell size, and be properly registered.
The stack orientation is the compass direction of the layers.
For geographic data, north or near-north should be at the top of the array, layer, screen, and printout.
There is a stack of base layers of images in the database
Creating stackable layers requires an input effort.
Once a stack of base layers is created, derivative layers can be generated.
These stacks may involve maps and images, and even stacked graph, or table data.