Scribd
Upload
2 views4 pages

CV Lab7

The document provides lab instructions for undergraduate students on Hough Transforms, focusing on line and circle detection using OpenCV functions. It includes code examples for detecting lines in an image with cv2.HoughLines and detecting circles with cv2.HoughCircles, along with questions to explore the effects of various parameters. Additionally, it contains a task to count coins in an image using the Hough Circle Transform, encouraging experimentation with parameters for optimal results.

Uploaded by

morteza karimi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views4 pages

CV Lab7

The document provides lab instructions for undergraduate students on Hough Transforms, focusing on line and circle detection using OpenCV functions. It includes code examples for detecting lines in an image with cv2.HoughLines and detecting circles with cv2.HoughCircles, along with questions to explore the effects of various parameters. Additionally, it contains a task to count coins in an image using the Hough Circle Transform, encouraging experimentation with parameters for optimal results.

Uploaded by

morteza karimi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Fundamentals of Computer Vision (Undergrad) - B.

Nasihatkon
Spring 2025

Lab Instructions - session 7


Hough Transforms

Part 1. Hough Line Transform

Detect lines in an image using the function cv2.HoughLines


File: hough_line.py
def draw_line(Img, rho, theta):
"""draws a line in an image 'Img' given 'rho' and 'theta'"""
a = np.cos(theta)
b = np.sin(theta)
x0 = a * rho
y0 = b * rho
x1 = int(x0 + 1000 * (-b))
y1 = int(y0 + 1000 * a)
x2 = int(x0 - 1000 * (-b))
y2 = int(y0 - 1000 * a)

cv2.line(Img, (x1, y1), (x2, y2), (0, 0, 255), 1)

Img = cv2.imread('sudoku.jpg')

G = cv2.cvtColor(Img, cv2.COLOR_BGR2GRAY) # -> grayscale

E = cv2.Canny(G, 100, 200) # find the edges

min_votes = 160 # minimum votes to be considered a line


distance_resolution = 1 # 1 pixel: resolution of the parameter "rho"
(distance to origin)
angle_resolution = np.pi / 180 # pi/180 radians: resolution (bin size) of
the parameter "theta"
L = cv2.HoughLines(E, distance_resolution, angle_resolution, min_votes)

# draw the lines


for [[rho, theta]] in L:
draw_line(Img, rho, theta)

cv2.imshow("E", E)
cv2.imshow("Img", Img)
cv2.waitKey(0)
cv2.destroyAllWindows()

●​ What happens by increasing or decreasing the parameter min_votes? Why?


●​ What is the effect of increasing and decreasing the distance_resolution
and angle_resolution parameters? Explain.
●​ If an image has faint lines (low contrast), should min_votes be increased or
decreased? Why?
Fundamentals of Computer Vision (Undergrad) - B. Nasihatkon
Spring 2025

●​ Would a very fine angle resolution (e.g., np.pi/1000) improve results? Why or
why not?
●​ What kind of lines (strong/weak edges) are prioritized when min_votes is
high?

Part 2: Hough Circle Transform


The goal is to detect the cups of coffee in the picture using cv2.HoughCircles

File: hough_circle.py
import numpy as np​
import cv2​
​
I = cv2.imread('coffee.jpg')​
​
G = cv2.cvtColor(I,cv2.COLOR_BGR2GRAY) # -> Grayscale​
G = cv2.GaussianBlur(G, (3,3), 0); # Gaussian blur​
​
canny_high_threshold = 200 ​
min_votes = 100 # minimum no. of votes to be considered as a circle​
min_centre_distance = 40 # minimum distance between the centers of detected circles​
resolution = 1 # resolution of parameters (centre, radius) relative to image resolution​
circles = cv2.HoughCircles(G,cv2.HOUGH_GRADIENT,
resolution,min_centre_distance,​
param1=canny_high_threshold,​
param2=min_votes,minRadius=0,maxRadius=100)​
​
for c in circles[0,:]:​
x = c[0] # x coordinate of the centre​
y = c[1] # y coordinate of the centre​
r = c[2] # radius​
​
# draw the circle​
cv2.circle(I,(x,y), r, (0,255,0),2)​
​
# draw the circle center ​
cv2.circle(I,(x,y),2,(0,0,255),2)​
​
cv2.imshow("I",I)​
cv2.waitKey(0)​
cv2.destroyAllWindows()
●​ Change the parameters of cv2.HoughCircles and see how each of them
affect detection.
●​ Can you explain why some circles disappear when min_votes is increased?
●​ What if you set it to a very large value (e.g., 100)?
●​ How does changing canny_high_threshold affect edge detection before circle
detection?
●​ How does setting minRadius=20 and maxRadius=50 change the output
compared to minRadius=0 and maxRadius=100?
Fundamentals of Computer Vision (Undergrad) - B. Nasihatkon
Spring 2025

Task1 : count the coins


You need to count the number of coins in the next image:

Write a piece of code to perform this task using a hough circle transform. Change the
file task1.py to perform the task. Play with the parameters until you get the desired
results.

File: task1.py
import numpy as np​
import cv2​
​
I = cv2.imread('coins.jpg')​
G = cv2.cvtColor(I,cv2.COLOR_BGR2GRAY)​
G = cv2.GaussianBlur(G, (5,5), 0);​
​
canny_high_threshold = 160​
min_votes = 30 # minimum no. of votes to be considered as a circle​
min_centre_distance = 40​
​
circles = np.array([[10,10]])​
​
for c in circles[0,:]:​
x = 100​
y = 100​
r = 40​
cv2.circle(I,(x,y), r, (0,255,0),2)​
print(circles.shape)​
n = 100​
font = cv2.FONT_HERSHEY_SIMPLEX​
cv2.putText(I,'There are %d coins!'%n,(400,40), font, 1,(255,0,0),2)​
​
cv2.imshow("I",I)​
cv2.waitKey(0)

●​ What happens by changing different parameters?


●​ The Hough transform can even detect the partially occluded coins. Why is this
the case?
Fundamentals of Computer Vision (Undergrad) - B. Nasihatkon
Spring 2025

References
●​ Hough Line Transform
●​ Hough Circle Transform

You might also like