Hint: Matlab has a lot of useful functions for this problem set.
"help Use the first image to create a set of training e's. Test on this
training image and calculate your accuracy:
Acc = (N - D - S - I) / N
Where N is the number of samples, D are the deletions (failed to
recognize the letter was present), S are the substitutions
(incorrectly recognized the letter), I are insertions (labeled
noise (i.e. something not a letter) as the particular letter).
Use the second image as an independent test (i.e. use the
same system that you developed for the first image on the second but
without using any examples from the second image to train the
system). What accuracy did you achieve?
Do the same thing with the third image. What was the accuracy?
Did your method generalize well? Try to use other features to help
recognize the letter. What did you use? How much did it improve the
accuracy?
Click Right Mouse button (and choose Save As)to download in
TIFF or in GIF, OR copy from
/net/www/classes/AY2002/cs4495_fall/ps/ps4/ps1-ex1.tiff
Click Right Mouse button (and choose Save As) to download in
TIFF or in GIF OR copy
from /net/www/classes/AY2002/cs4495_fall/ps/ps4/ps1-ex2.tiff
Click Right Mouse button (and choose Save As) to download in
TIFF or in GIF OR copy
from /net/www/classes/AY2002/cs4495_fall/ps/ps4/ps1-ex3.tiff
This problem set requires a little bit more thinking and algorithm
construction than the previous problem sets. Therefore it is essential
that you turn in the following in your README/index.html file
(You
will not get full credit otherwise):
Your output for each vowel should be three images in which the
detected letters are highlighted.
- the assumptions and
- the weaknesses of the particular technique(s) employed.
- How does your program fail and why?
- How would you extend the program to make it more general?
- What have you learned?
You will submit your work electronically (no paper). Instead, you will submit an HTML file. Create your own web page in the spirit of the template provided for PS2. The format of the page is up to you. Feel free to add as many result images, intermediate results images, and result images for different values of the control parameters as you think is necessary to demonstrate your work. However, the page should contain at least the following:
- your name on top of the page,
- links to ALL of your matlab code,
- links to your result images converted to JPG format.
Your matlab code should be reasonably commented if you want us to understand it and give you credit/comments. Also include test scripts for each of the two subproblems: test_e_only.m and test_vowels.m. These files should demonstrate your work for part 1 (letter e only), and part 2 (all vowels) of this problem set, respectively. test_vowels.m will be used to determine the winner for part 3. The scripts should:
- load your test images,
- call your other matlab functions that do the required image processing,
- show all results on the screen using multiple figures, and
- for your final result images color the pixels of the detected letters in some distinct color so we know that your code is indeed detecting them. Use different colors for different letter (see example below).
- save all result images in JPG format with appropriate names that can be loaded from the web page.
To submit your work do the following:
- On a UNIX platform create a directory named after your CoC login_name
- Put ALL your matlab files, the three test scripts, the original images, and all result images in that directory.
- Also put a README file that explains any special needs of your code, and lists any KNOWN BUGS/FEATURES.
For example if your code is running very slow try to explain why. If something is not working explain why.
Also remember to put the required explanations of your code as described above. - Create an HTML page according to the instructions given above. Save it as index.html in the same directory.
- CD to the parent directory.
- Create a tar archive using:
- tar cvf login_name.tar login_name
- Gzip the tar file using:
- gzip login_name.tar
- E-mail the login_name.tar.gz file to cs7495@cc.gatech.edu