Scribd
Upload
2 views

CSE463 Lab Assignment 2

The CSE463 Lab Assignment 2 requires students to complete various tasks involving image processing techniques, including convolution, padding, noise reduction, edge detection, and histogram equalization using a total of 7 images. Students must upload their images and code outputs to a Google Drive folder and share the link in the submission form by the deadline of March 8, 2025. Each task involves applying specific filters and kernels to analyze and describe the effects on the images.

Uploaded by

leading.armadillo.sajp
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views

CSE463 Lab Assignment 2

The CSE463 Lab Assignment 2 requires students to complete various tasks involving image processing techniques, including convolution, padding, noise reduction, edge detection, and histogram equalization using a total of 7 images. Students must upload their images and code outputs to a Google Drive folder and share the link in the submission form by the deadline of March 8, 2025. Each task involves applying specific filters and kernels to analyze and describe the effects on the images.

Uploaded by

leading.armadillo.sajp
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

CSE463 Lab Assignment 2

Deadline: March 8, 2025, 11:59PM

Submission instructions:

You will require a total of 7 different images in this assignment. The numbers are
mentioned in square brackets.

● Open a folder in Google Drive and name it as- ID_Name_Lab2


a. Upload the images you used (number them the way [..] is numbered, eg.
Image 1, Image 2, etc.)
b. Upload a PDF consisting of the codes & screenshots of the outputs of each
code [Optional]
c. Ensure that you submit an ipynb file containing all the outputs of your
results.
● Share the link of the folder in the submission form. (Make sure the folder is
accessible(Anyone with the link)
● Submission Link: https://forms.gle/mq7xB2XUoT6iwp5b8

Step 1: Exploring Basic Convolution and Custom Kernels

1. Apply Convolution with a Simple Kernel


○ Task: Given a grayscale image[1], apply a basic 3x3 identity kernel to the
image using a 2D convolution function (e.g., cv2.filter2D). Observe and
describe the output in 1-2 sentences.
2. Custom Kernel Design
○ Task: Design a 3x3 custom kernel to create a sharpening effect on an
image[2]. Apply the kernel to the image and describe how this kernel
affects the overall clarity and contrast of the image. [Opposite of blurring]

Step 2: Understanding Padding and Its Effects on Convolution

3. Experiment with Different Padding Techniques


○ Task: Apply the sharpening kernel created in Step 1 to an image[3] with
three different padding types: constant (zero-padding), reflect padding,
and same padding.

Step 3: Filtering for Noise Reduction and Smoothing


4. Adding Noise and Applying an Average Filter

○ Task: Add Gaussian noise to an image[4]. Apply a 5x5 average filter


using cv2.blur and observe the changes in noise level and overall
appearance of the image. Explain in 1-2 sentences.

5. Gaussian Blur for Smoothing

○ Task: Apply a Gaussian blur to the noisy image created in Part 4 using a
5x5 Gaussian kernel (cv2.GaussianBlur). Experiment with different
standard deviations (sigma values) and observe how the level of smoothing
changes. Explain in 1-2 sentences.

Step 4: Edge Detection and Gradient Calculation

6. Applying Laplacian Filter for Edge Detection

Task: Use the Laplacian filter to detect edges in a clear image[5]. Apply the
Laplacian filter on the grayscale image, then visualize the result.

7. Estimating Horizontal and Vertical Gradients

Task: To estimate the edges in horizontal and vertical directions, apply two simple
gradient kernels:

● For vertical edges, use:

● For horizontal edges, use:

Apply each kernel separately to an image[6] and visualize the results to observe
the types of edges each captures. Explain in 1-2 sentences. (Just one more 😐)

Step 5: Image Enhancement through Histogram Equalization


8. Histogram Equalization for Contrast Enhancement

○ Task 1: Load a low-contrast grayscale image[7] and apply histogram


equalization. Observe and describe in 1-2 sentences how the contrast
improves.
○ Task 2: Apply histogram equalization three times on the original image[7]
and describe any diminishing effects or artifacts that appear. Explain in 1-2
sentences (One last time 🙂)

You might also like