OpenCV Python Library Cheatsheet

Computer Vision is a rapidly evolving field that involves teaching machines to interpret and understand visual information from the world around us. OpenCV (Open Source Computer Vision Library) plays a pivotal role in this domain by providing a plethora of tools and functions for image and video processing. To help both beginners and seasoned practitioners navigate through the rich features of OpenCV, we present a comprehensive cheatsheet that covers essential concepts and functions.

1. Installation

Before diving into the OpenCV world, make sure you have it installed. You can use the following command for Python:

pip install opencv-python

For the latest version:

pip install opencv-python-headless

2. Importing OpenCV

import cv2

3. Reading and Displaying Images

# Read an image from file
image = cv2.imread('path/to/image.jpg')

# Display the image
cv2.imshow('Image', image)
cv2.waitKey(0)
cv2.destroyAllWindows()

4. Reading and Displaying Videos

# Read a video from file or camera
cap = cv2.VideoCapture('path/to/video.mp4')  # or 0 for default camera

while True:
    ret, frame = cap.read()

    # Display the video frame
    cv2.imshow('Video', frame)

    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()

5. Basic Image Operations

Grayscale Conversion

gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

Resizing

resized_image = cv2.resize(image, (width, height))

Blurring

blurred_image = cv2.GaussianBlur(image, (kernel_size, kernel_size), 0)

6. Drawing on Images

Lines

cv2.line(image, (x1, y1), (x2, y2), (0, 255, 0), thickness)

Rectangles

cv2.rectangle(image, (x, y), (x + width, y + height), (0, 255, 0), thickness)

Circles

cv2.circle(image, (center_x, center_y), radius, (0, 255, 0), thickness)

7. Image Thresholding

ret, thresholded_image = cv2.threshold(gray_image, threshold_value, max_value, cv2.THRESH_BINARY)

8. Contour Detection

contours, hierarchy = cv2.findContours(thresholded_image, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

9. Face Detection (Using Haarcascades)

face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
faces = face_cascade.detectMultiScale(gray_image, scaleFactor, minNeighbors)

10. Image Transformation

Affine Transformation

M = cv2.getAffineTransform(pts1, pts2)
transformed_image = cv2.warpAffine(image, M, (width, height))

Perspective Transformation

M = cv2.getPerspectiveTransform(pts1, pts2)
perspective_transformed_image = cv2.warpPerspective(image, M, (width, height))

Conclusion

This cheatsheet provides a quick reference for essential OpenCV operations. Keep in mind that this is just the tip of the iceberg, and OpenCV offers many more advanced features. Experiment with these functions and explore the vast possibilities that OpenCV brings to the world of Computer Vision.

FAQ