Description

TAs: Alankar Kotwal, Martin Li, Dinesh Reddy Narapureddy, Shumian Xin, Ye Yuan, and Xian Zhou
This and future assignments will assume that you are using Python 3+ and the latest Scipy libraries (numpy, matplotlib, etc). The recommended approach is to use the Anaconda Python 3 distribution .
1 Color Channel alignment
You have been given the red, green, and blue channels of an image that were taken separately using an old technique that captured each color on a separate piece of glass2.
The easiest way to do this is to exhaustively search over a window of possible displacements for the different channels (you can assume the displacement will be between -30 and 30 pixels). Score the alignment from each possible displacement with some heuristic and choose the best alignment using these scores. You can use the following metrics:

Figure 1: Combining the red, green, and blue channels without shifting

Figure 2: Aligned image
1. Sum of Squared Differences (SSD)
This metric tries to compare the distance between two vectors, hence we can use this to compare image pixel intensity differences. For two vectors u, v of length N, this metric is defined as
N
SSD(u,v) = X(u[i] − v[i])2
i=1
(essentially like the Euclidean distance metric).
2. Normalized Cross Correlation (NCC)
This metric compares normalized unit vectors using a dot product, that is for
vectors u,v,
u v
NCC(u,v) = ·
kuk kvk
Implement an algorithm which finds the best alignment of the three channels to produce a good RGB image; it should look something like Figure 2. Try to avoid using for loops when computing the metrics (SSD or NCC) for a specific offset. Some starter code is given in script1.py and alignChannels.py. Save your result as rgb output.jpg in the results folder. Also include the resulting image in the writeup.
2 Image warping
We will be writing code that performs affine warping on an image. Code has been provided to get you started. In the following sections you will be adding your own code.
2.1 Example code
The file script2.py contains example code which demonstrates basic image loading and displaying as well as the behavior of an image warping function. Try running script2. You should see something like Figure 3. This script calls the function warp in warpA_check.py, which is simply a wrapper for scipy’s own function scipy.ndimage.affine_transform.
2.2 Affine warp
You will write your own function in warpA.py, that should give the same output as the function in warpA_check.py. First we’ll walk through what you need to know about affine transformations.
Figure 3: See script2.py

An affine transform relates two sets of points:
p t (1)

| {z }
L
where pisource and piwarped denote the 2D coordinates (e.g., p ) of the i-th point in the source space and destination (or warped) space respectively, L is a 2 × 2 matrix representing the linear component, and t is a 2 × 1 vector representing the translational component of the transform.
To more conveniently represent this transformation, we will use homogeneous coordinates, i.e., and piw will now denote the 2D homogeneous coordinates (e.g., p
), where “≡” denotes equivalence up to scale, and the transform becomes:

pid ≡pis (2)
| {z }
A
• Implement a function that warps image im using the affine transform A: warp_im = warp(im, A, output_shape)
Inputs: im is a grayscale double typed Numpy matrix of dimensions height×width×1 , A is a 3 × 3 non-singular matrix describing the transform ( Apisource), and output_size=[out_height,out_width]; of the warped output image.
Outputs: warp_im of size out_size(1) × out_size(2) is the warped output image. The coordinates of the sampled output image points piwarped should be the rectangular range (0,0) to (width − 1,height − 1) of integer values. The points pisource must be chosen such that their image, Ap , transforms to this rectangle.
Implementation-wise, this means that we will be looking up the value of each of the destination pixels by sampling the original image at the computed pisource. (Note that if you do it the other way round, i.e., by transforming each pixel in the source image to the destination, you could get “holes” in the destination because the mapping need not be 1 to 1). In general, the transformed values p will not lie at integer locations and you will therefore need to choose a sampling scheme; the easiest is nearest-neighbor sampling (something like, round(p ). You should be able to implement this without using for loops (one option is to use numpy.meshgrid and Numpy’s multidimensional indexing), although it might be easier to implement it using loops first. Save the resulting image in results/transformed.jpg. Also include the resulting image in the write up.
You should check your implementation to make sure it produces the same output as the warp function provided in warpA check.py (for grayscale or RGB images). Obviously the purpose of this exercise is practicing Python/Scipy by implementing your own function without using anything like scipy.ndimage.affine_transform.
3 Getting started with OpenCV
We already have the latest version of Anaconda and conda running on your systems. We would recommend creation of a virtual environment with the with python version 3.7 installed. Using anaconda you can use the following command to create a new virtual environment: conda create -n cv_assignment python=3.7
The following command will help you in installing the latest build of OpenCV. conda install -c conda-forge opencv
Check your OpenCV installation in python by typing the following command:
import cv2
If the above works without any kind of errors, you are good to go. If errors persist, try googling, stack overflowing and make it work. This step is going to save you a lot of time in future.
4 Submission and Deliverables
We use Gradescope for homework submission and grading. Please submit your code as a zip file to
Gradescope HW0_Code of the form <andrewID>.zip, and writeup with the name <andrewID>.pdf containing two result images to Gradescope HW0_Writeup. Running the scripts in the code folder should generate the desired results in the results folder, as explained in each of the section. You should not include the data/ folder in your submission. An example submission zip file should be: sigbovik.zip:
• code/
– alignChannels.py
– script1.py
– script2.py
– warpA.py
– warpA_check.py
– any other files your code needs to run
• results/
– rgb_output.jpg
– transformed.jpg