OpenCV

OpenCV ¶

OpenCV (Open Source Computer Vision) is a library of programming functions mainly aimed at real-time computer vision. Originally developed by Intel, it was later supported by Willow Garage then Itseez (which was later acquired by Intel). OpenCV currently supports wide variety of languages like, C++, Python, Java, etc.

Installation ¶

Please refer to these articles for installation of OpenCV on your computer.

Windows Installation Instructions: https://opencv-python-tutroals.readthedocs.io/en/latest/py_tutorials/py_setup/py_setup_in_windows/py_setup_in_windows.html#install-opencv-python-in-windows
Mac Installation Instructions (High Sierra): https://medium.com/@nuwanprabhath/installing-opencv-in-macos-high-sierra-for-python-3-89c79f0a246a
Linux Installation Instructions (Ubuntu 18.04): https://www.pyimagesearch.com/2018/05/28/ubuntu-18-04-how-to-install-opencv

Here we will be focusing on Python implementation of OpenCV ¶

  1# Reading image in OpenCV
  2import cv2
  3img = cv2.imread('cat.jpg')
  4
  5# Displaying the image
  6# imshow() function is used to display the image
  7cv2.imshow('Image', img)
  8# Your first argument is the title of the window and second parameter is image
  9# If you are getting an error, Object Type None, your image path may be wrong. Please recheck the path to the image
 10cv2.waitKey(0)
 11# waitKey() is a keyboard binding function and takes an argument in milliseconds. For GUI events you MUST use waitKey() function.
 12
 13# Writing an image
 14cv2.imwrite('catgray.png', img)
 15# The first argument is the file name and second is the image
 16
 17# Convert image to grayscale
 18gray_image = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
 19
 20# Capturing Video from Webcam
 21cap = cv2.VideoCapture(0)
 22# 0 is your camera, if you have multiple cameras, you need to enter their id
 23while True:
 24    # Capturing frame-by-frame
 25    _, frame = cap.read()
 26    cv2.imshow('Frame', frame)
 27    # When user presses q -> quit
 28    if cv2.waitKey(1) & 0xFF == ord('q'):
 29        break
 30# Camera must be released
 31cap.release()
 32
 33# Playing Video from file
 34cap = cv2.VideoCapture('movie.mp4')
 35while cap.isOpened():
 36    _, frame = cap.read()
 37    # Play the video in grayscale
 38    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
 39    cv2.imshow('frame', gray)
 40    if cv2.waitKey(1) & 0xFF == ord('q'):
 41        break
 42cap.release()
 43
 44# Drawing The Line in OpenCV
 45# cv2.line(img, (x,y), (x1,y1), (color->r,g,b->0 to 255), thickness)
 46cv2.line(img, (0, 0), (511, 511), (255, 0, 0), 5)
 47
 48# Drawing Rectangle
 49# cv2.rectangle(img, (x,y), (x1,y1), (color->r,g,b->0 to 255), thickness)
 50# thickness = -1 used for filling the rectangle
 51cv2.rectangle(img, (384, 0), (510, 128), (0, 255, 0), 3)
 52
 53# Drawing Circle
 54# cv2.circle(img, (xCenter,yCenter), radius, (color->r,g,b->0 to 255), thickness)
 55cv2.circle(img, (200, 90), 100, (0, 0, 255), -1)
 56
 57# Drawing Ellipse
 58cv2.ellipse(img, (256, 256), (100, 50), 0, 0, 180, 255, -1)
 59
 60# Adding Text On Images
 61cv2.putText(img, "Hello World!!!", (x, y), cv2.FONT_HERSHEY_SIMPLEX, 2, 255)
 62
 63# Blending Images
 64img1 = cv2.imread('cat.png')
 65img2 = cv2.imread('openCV.jpg')
 66dst = cv2.addWeighted(img1, 0.5, img2, 0.5, 0)
 67
 68# Thresholding image
 69# Binary Thresholding
 70_, thresImg = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY)
 71# Adaptive Thresholding
 72adapThres = cv2.adaptiveThreshold(img, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 11, 2)
 73
 74# Blur Image
 75# Gaussian Blur
 76blur = cv2.GaussianBlur(img, (5, 5), 0)
 77# Median Blur
 78medianBlur = cv2.medianBlur(img, 5)
 79
 80# Canny Edge Detection
 81img = cv2.imread('cat.jpg', 0)
 82edges = cv2.Canny(img, 100, 200)
 83
 84# Face Detection using Haar Cascades
 85# Download Haar Cascades from https://github.com/opencv/opencv/blob/master/data/haarcascades/
 86import cv2
 87import numpy as np
 88
 89face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
 90eye_cascade = cv2.CascadeClassifier('haarcascade_eye.xml')
 91
 92img = cv2.imread('human.jpg')
 93gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
 94
 95faces = face_cascade.detectMultiScale(gray, 1.3, 5)
 96for x, y, w, h in faces:
 97    # Draw a rectangle around detected face
 98    cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
 99    roi_gray = gray[y : y + h, x : x + w]
100    roi_color = img[y : y + h, x : x + w]
101    eyes = eye_cascade.detectMultiScale(roi_gray)
102    for ex, ey, ew, eh in eyes:
103        # Draw a rectangle around detected eyes
104        cv2.rectangle(roi_color, (ex, ey), (ex + ew, ey + eh), (0, 255, 0), 2)
105
106cv2.imshow('img', img)
107cv2.waitKey(0)
108
109cv2.destroyAllWindows()
110# destroyAllWindows() destroys all windows.
111# If you wish to destroy specific window pass the exact name of window you created.

OpenCV ¶

Installation ¶

Here we will be focusing on Python implementation of OpenCV ¶

Further Reading: ¶