Project #1 Edge Detection

CSE 473/573 – Computer Vision and Image Processing

Project #1

1. Edge Detection [35 points]
The goal of this task is to experiment with two commonly used edge detection operator, i.e.,
Prewitt operator and Sobel operator, and familiarize you with tricks, e.g., padding, commonly
used by computer vision practitioners. Specifically, the task is to detect edges in a given image,
which is named “proj1-task1.jpg” and is stored in “./data/”.
You are required to implement all the functions that are labelled with “# TODO” in “task1.py”
and “utils.py”. In “task1.py” and “utils.py”, we not only provide hints to you, but also provide
utility functions that could be used as building blocks for you to complete this task. Therefore,
you only need to write about 40 lines of code. Your code should be able to generate images that
are identical to those stored in “./results/” (Note that images in “./results/” are provide to you
for reference only. At test time, a different image will be used as input and the correct output
images will be different from those already stored in “./results/”).
Comment the lines “raise NotImplementedError” instead of deleting them, when you implement
the functions labelled with “# TODO”.
2. Character Detection [65 points]
The goal of this task is to experiment with template matching algorithms. Specifically, the task
is to find a specific character (or set of characters) in a given image, which is named “proj1-
task2.pgm” and is store the results in “./data/”. You are required to implement a function named
“detect” in “task2.py”, which detects a character in an image. The function “detect” takes a
given image and a given template (that you will implement) that contains a character as input
and returns the coordinates (i.e., coordinates of the top-left pixel) of the character contained in
the template.
This task is composed of the following three sub tasks:
• Detect character “a” (lower case “a”). [3 point]
• Detect character “b” (lower case “b”). [3 point]
• Detect character “c” (lower case “c”). [3 point]
You need to customize your own templates. The templates containing characters “a”, “b”
and “c” should be named as “a.pgm”, “b.pgm” and “c.pgm” respectively. They should be
stored in “./data/”. Note that only ONE template will be used to detect all instance
of a particular (lowercase) character, but it can be scaled or rotated, for example, in
your program.
Evaluation
The F1 measure is the harmonic mean of Precision and Recall,
with Precision being the # of true positives / # one says are
positive, and the recall is the # of true positive / # of positives that
exist.
The FI will be used as the metric. You are not expected to get all
the characters, but so you can assume that if you get an F1 measure
0.6 you will get full credit
Hints:
• You may consider designing a grayscale template that emphasizes some pixels more
than others.
• You may consider combining several “exact” template in an appropriate way to be able
to recognize different variations of the 3 lower case characters in question.
• Once you read in your base template your program can further manipulate it for, for
example, size.
• After you have a base system, you may also experiment with templates which only
contain edges for template matching (detecting edges in both the image and the
template). This might give better resultsthan using the original image and template
for template matching, as edges only preserve the scales and shapes of characters,
which are important for template matching. Distracting information, e.g., colors and
fonts, of the characters are partially eliminated in the image and template that only
contain edges. (Functions that we provide to you in “utils.py” and functions that
you implement in “task1.py” could be used to detect edges.)
• PGM images are a simple raw representation of any image, and additional
information can be found here: http://netpbm.sourceforge.net/doc/pgm.html
Project Guidelines
• Do not modify the code provided.
• Do not use any API provided by opencv (cv2) and numpy (np) in your code (except
“np.sqrt()”, “np.zeros()”, “np.ones()”, “np.multiply()”, “np.divide()”, “cv2.imread()”,
“cv2.imshow()”, “cv2.imwrite()”, and “cv2.resize()”).
• Do not import any additional libraries (function, module, etc.) except native Python
packages, e.g., pdb, os, sys.
• You will upload your project to be UBLearns and to Autograder
(https://autograder.cse.buffalo.edu/), so please try both early in the process to make sure you understand
•
how they work.
Ask for the TA’s approval if you would like to use a programming language other than
Python, C, C++, C# and Java.