Scribd
Upload
19 views6 pages

AliCV A2

Uploaded by

Ali Razzaq

The document is an assignment submission that includes code to perform computer vision tasks on an image, including: 1) Converting the image to grayscale, applying thresholding, erosion, and dilation. 2) Smoothing and blurring the grayscale image using an averaging filter. 3) Detecting edges using Sobel kernels to calculate gradients in the x and y directions and taking the gradient magnitude. 4) Displaying the original and processed images.

Copyright:

© All Rights Reserved
Download as DOCX, PDF, TXT or read online from Scribd
Download as docx, pdf, or txt

AliCV A2

Uploaded by

Ali Razzaq
19 views6 pages
The document is an assignment submission that includes code to perform computer vision tasks on an image, including: 1) Converting the image to grayscale, applying thresholding, erosion, and dilation. 2) Smoothing and blurring the grayscale image using an averaging filter. 3) Detecting edges using Sobel kernels to calculate gradients in the x and y directions and taking the gradient magnitude. 4) Displaying the original and processed images.

Original Description:

computer visio

Original Title

AliCV_A2 (1)

Copyright

© © All Rights Reserved

Available Formats

DOCX, PDF, TXT or read online from Scribd

Did you find this document useful?

Is this content inappropriate?

The document is an assignment submission that includes code to perform computer vision tasks on an image, including: 1) Converting the image to grayscale, applying thresholding, erosion, and dilation. 2) Smoothing and blurring the grayscale image using an averaging filter. 3) Detecting edges using Sobel kernels to calculate gradients in the x and y directions and taking the gradient magnitude. 4) Displaying the original and processed images.

Copyright:

© All Rights Reserved
Download as DOCX, PDF, TXT or read online from Scribd
Download as docx, pdf, or txt
19 views6 pages

AliCV A2

Uploaded by

Ali Razzaq
The document is an assignment submission that includes code to perform computer vision tasks on an image, including: 1) Converting the image to grayscale, applying thresholding, erosion, and dilation. 2) Smoothing and blurring the grayscale image using an averaging filter. 3) Detecting edges using Sobel kernels to calculate gradients in the x and y directions and taking the gradient magnitude. 4) Displaying the original and processed images.

Copyright:

© All Rights Reserved
Download as DOCX, PDF, TXT or read online from Scribd
Download as docx, pdf, or txt
You are on page 1/ 6

Name: ALI RAZZAQ

# ID: F2020266307

Assignment: 2 Computer Vision

Question # 1:
Step 1: Importing Libraries and Upload the Image

from google.colab import files


from PIL import Image
import numpy as np
import matplotlib.pyplot as plt

# Upload an image file


uploaded = files.upload()
filename = next(iter(uploaded)) # Get the filename of the uploaded
image

<IPython.core.display.HTML object>

Saving images.jpeg to images.jpeg

Step 2: Image to a Grayscale Array

# Open the original image


with Image.open(filename) as img:
# Convert to grayscale
gray = img.convert('L')
img_array = np.array(gray)

# Display both original and grayscale images


fig, axs = plt.subplots(1, 2, figsize=(12, 6))

# Display the original image


axs[0].imshow(img) # img is the original image
axs[0].set_title('Original Image')
axs[0].axis('off') # Turn off axis numbers and labels

# Display the grayscale image


axs[1].imshow(img_array, cmap='gray')
axs[1].set_title('Grayscale Image')
axs[1].axis('off') # Turn off axis numbers and labels

plt.show()
Step 4: Applying Thresholding

# Thresholding
thresh_value = 128
height, width = img_array.shape
thresholded = np.zeros((height, width), dtype=np.uint8)

for i in range(height):
for j in range(width):
thresholded[i, j] = 255 if img_array[i, j] > thresh_value else
0

Step 5: Applying Erosion

# Manual Erosion
kernel_size = 3
eroded = np.zeros((height, width), dtype=np.uint8)

for i in range(height - kernel_size + 1):


for j in range(width - kernel_size + 1):
if np.all(thresholded[i:i + kernel_size, j:j + kernel_size] ==
255):
eroded[i + kernel_size // 2, j + kernel_size // 2] = 255

Step 6: Applying Dilation

# Dilation
dilated = np.zeros((height, width), dtype=np.uint8)

for i in range(height - kernel_size + 1):


for j in range(width - kernel_size + 1):
if np.any(thresholded[i:i + kernel_size, j:j + kernel_size] ==
255):
dilated[i + kernel_size // 2, j + kernel_size // 2] = 255

Step 7: Displaying the Processed Images


# Display the original, thresholded, eroded, and dilated images
fig, axs = plt.subplots(1, 4, figsize=(20, 5))
axs[0].imshow(img, cmap='gray')
axs[0].set_title('Original Image')
axs[1].imshow(thresholded, cmap='gray')
axs[1].set_title('Thresholded Image')
axs[2].imshow(eroded, cmap='gray')
axs[2].set_title('Eroded Image')
axs[3].imshow(dilated, cmap='gray')
axs[3].set_title('Dilated Image')
plt.show()

Question # 2:
Step 1: Set Up Google Colab and Mount Google Drive

from google.colab import drive


drive.mount('/content/drive')

Drive already mounted at /content/drive; to attempt to forcibly


remount, call drive.mount("/content/drive", force_remount=True).

Step 2: Read the Image and Convert to Grayscale

from PIL import Image


import numpy as np
import matplotlib.pyplot as plt

img_path = ('/content/drive/My Drive/images.jpeg')

# Open the image and convert to grayscale


with Image.open(img_path) as img:
grayscale = img.convert('L')
image_array = np.array(grayscale)

# Display both original and grayscale images


fig, axs = plt.subplots(1, 2, figsize=(12, 6))

# Display the original image


axs[0].imshow(img) # img is the original image
axs[0].set_title('Original Image')
axs[0].axis('off') # Turn off axis numbers and labels

# Display the grayscale image


axs[1].imshow(image_array, cmap='gray')
axs[1].set_title('Grayscale Image')
axs[1].axis('off') # Turn off axis numbers and labels

plt.show()

Step 3: Apply Smoothing and Blurring

# kernel size
k = 3

height, width = image_array.shape


blurred_image = np.zeros((height, width), dtype=np.uint8)

# Averaging filter
for i in range(k, height - k):
for j in range(k, width - k):
# Average of neighborhood
sum_pixels = 0
for di in range(-k, k + 1):
for dj in range(-k, k + 1):
sum_pixels += image_array[i + di, j + dj]
blurred_image[i, j] = sum_pixels // (kernel_size *
kernel_size)

# Display both original and blurred image


fig, axs = plt.subplots(1, 2, figsize=(12, 6))

# Display the original image


axs[0].imshow(img) # img is the original image
axs[0].set_title('Original Image')
axs[0].axis('off') # Turn off axis numbers and labels
# Display the grayscale image
axs[1].imshow(blurred_image, cmap='gray')
axs[1].set_title('Blured image')
axs[1].axis('off') # Turn off axis numbers and labels

plt.show()

Step 4: Edge Detection

# Sobel kernels
sobel_x = np.array([[-1, 0, 1], [-2, 0, 2], [-1, 0, 1]])
sobel_y = np.array([[-1, -2, -1], [0, 0, 0], [1, 2, 1]])

# Initialize the gradient arrays


gradient_x = np.zeros_like(blurred_image, dtype=np.int32)
gradient_y = np.zeros_like(blurred_image, dtype=np.int32)

# Compute gradient in x and y direction


for i in range(1, height - 1):
for j in range(1, width - 1):
region = blurred_image[i - 1:i + 2, j - 1:j + 2]
gradient_x[i, j] = np.sum(region * sobel_x)
gradient_y[i, j] = np.sum(region * sobel_y)

# Magnitude of gradients
gradient_magnitude = np.sqrt(gradient_x**2 + gradient_y**2)
gradient_magnitude = np.clip(gradient_magnitude, 0,
255).astype(np.uint8)

# Display both original image and edge-detected image


fig, axs = plt.subplots(1, 2, figsize=(12, 6))

# Display the original image


axs[0].imshow(img) # img is the original image
axs[0].set_title('Original Image')
axs[0].axis('off') # Turn off axis numbers and labels
# Display the Edge Detection
plt.imshow(gradient_magnitude, cmap='gray')
plt.title('Edge Detected Image')
plt.axis('off')
plt.show()

You might also like