






G6-Final-Report
PRESENTING RESULTS FOR GROUP 6
Group Members : James Case, Dietz Ellis
Philip: Code, picture capture
Dietz: Picture capture, Picture Selection
Final Results

Here we have a result image. You can see black spots at the bottom of the metro sign. This happened because those pixels were pure black in the color image and thus the vectors here have ambiguous direction.
What IMAGES were used

This is the image, which controlled the intensity for the resultant image. We have found that the human visual system is much more sensitive to intensity than it is to color. Because of this it seams to work well to use images, which have subtle variations in intensity and are pattern like.

This is the image, which was used as the color for the resultant image. Because the human visual system seams to be less adept at reading variations in color, it seams to work well to use images with very different and sharply defined color segments as in this one.
DESCRIPTION
What were the goals?
n our project we attempted to create a method of conveying two meanings in a single photo. We do this by using the intensity values from one source image and the color values from a second in a resultant image. Black and white photography is nothing more than the intensity information from a scene. We simply use the unused variable of color to encode more information.
How stock footage was captured
With hand held digital cameras.
Other Material that was used (acquired!)
None
Tools used and HOW?
The Gimp to convert .jpg files into .ppm files
Code or scripting (include as an attachment OR link)
Here is a link to the code we wrote for this assignment. It is not meant to be a production program, simply an experiment. It ought to compile in windows. To use it you must use two source images with the same dimensions. They must be in the same directory as the compiled program and be called Darkness.ppm and Color.ppm. They must be ASCII ppm images raw will not work. After the program finishes a file called Output.ppm will be generated.
main.cpp
What worked?
With the right Images it the viewer is able to identify both images. It also works well as an abstract affect.
What was unexpected?
We had not thought about what would happen with black pixels in the color image sicne they have ambiguous direction.
What did NOT work?
The black pixel thing. It was dificult to see the color image sometimes.
How would you fix it?
The black pixel thing could be fixed by averaging the direction of the neigbhoring pixels and using this value.
Thoughts for the future of this effort?
Make it so you can tune the percentage of affect both in color and intensity the two images have on the resulting image. So one image could have 80% affect on the color and 20% on the intensity.
Other detials and DOCUMENTATION.
None
Links to this Page
- CP 2005 Groups last edited on 25 July 2005 at 6:02 am by 145.Red-81-32-187.pooles.rima-tde.net
- OLD Projects Page last edited on 25 July 2005 at 5:56 am by 145.Red-81-32-187.pooles.rima-tde.net