Python for Image Recognition - OpenCV

OpenCV is an open-source image recognition library.
It is used for machine learning, computer vision and image processing. You can extract the most out of OpenCV when integrated with powerful libraries like Numpy and Pandas.

INSTALLATION PYTHON 3.X

Open Terminal/Command Prompt and type :
~ pip install opencv-python

GETTING STARTED (HOW TO READ IMAGES)

1.Open PyCharm.
2.Import cv2.
3.Paste a test image in the directory.
4.Create variable to store image using imread() function.
5. Display the image using imshow() function.
6. Add a delay using a waitkey() function.

1
2
3
4
5
6
import cv2
# LOAD AN IMAGE USING 'IMREAD'
img = cv2.imread("Resources/lena.png")
# DISPLAY
cv2.imshow("Lena Soderberg”, img)
    cv2.waitKey(0)

PLAYING VIDEO USING VideoCapture() FUNCTION

1. Open PyCharm.

2. Import cv2.

3. Paste a test video in the directory.

4. Create variable to store video using VideoCapture() function.

5. Create an infinite while loop to display each frame of the video continuously.

6. Display the video using imshow() function.

7. Add a delay using a waitkey() function.

1
2
3
4
5
6
7
8
9
import cv2
frameWidth = 640
frameHeight = 480
cap = cv2.VideoCapture("Resources/test_ video.mp4")
while True:
  success, img = cap.read()
img = cv2.resize(img, (frameWidth, frameHeight))
cv2.imshow("Result", img)
break

ACCESSING LIVE FEED FROM WEBCAM

1. Open PyCharm.

2. Import cv2.

3. Create variable to store video using VideoCapture() function.

4. Pass parameter 0 in VideoCapture(0) to access webcam.

5. Create an infinite while loop to display each frame of the webcam’s video continuously.

6. Display the live feed using imshow() function.

7. Add a delay of infinity using waitKey(0).

1
2
3
4
5
6
7
8
9
10
11
import cv2
Width = 640
Height = 480
cap = cv2.VideoCapture(0)
cap.set(3, frameWidth)
cap.set(4, frameHeight)
cap.set(10, 150)
while True:
  success, img = cap.read()
cv2.imshow("Result", img)
break

FUNCTIONS OF OPENCV

Converting image to grayscale

1. Open PyCharm.

2. Import cv2.

3. Create variable to store image using imread() function.

4. To convert to grayscale use cv2.cvtColor() function

5. Pass the parameter image location and COLOR_BGR2GRAY to convert.

1
2
3
4
5
import cv2
img = cv2.imread("Resources/lena.png")
imgGray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cv2.imshow("Gray Image", imgGray)
cv2.waitKey(0)

Edge detection

1. Open PyCharm.

2. Import cv2.

3. Create variable to store image using imread() function.

4. To detect edge use cv2.Canny() function

5. Pass the parameter image location and threshold to convert.

1
2
3
4
5
6
import cv2

img = cv2.imread("Resources/lena.png")
imgCanny = cv2.Canny(img, 150, 200)
cv2.imshow("Canny Image”, imgCanny)
    cv2.waitKey(0)

CROPPING IMAGE

1. Import numpy and cv2.

2. Create two variables to store the height and width of the image.

3. Create two numpy arrays to store the coordinates.

4. First array - store the coordinates of the image to be cropped.

5. Second array - store the coordinates of the complete image.

6. Crop the image using getPerspective() and wrapPerspective() function.

1
2
3
4
5
6
7
8
9
10
11
image = cv2.imread("Assets/cards.jpg")

width, height = 250, 350
point1 = np.float32([[111, 219], [287, 188], [154, 482], [352, 440]])
point2 = np.float32([[0, 0], [width, 0], [0, height], [width, height]])
matrix = cv2.getPerspectiveTransform(point1, point2)
Output = cv2.warpPerspective(image, matrix, (width, height))

cv2.imshow("Image”, image)
    cv2.imshow("Output”, Output)
      cv2.waitKey(0)

FACE DETECTION

1. Open PyCharm.

2. Import cv2.
   3.Create a variable to store cascade classifier (to learn more about cascade classifier click here.

3. Convert image to greyscale using cv2.cvtColor() function.

4. Detect face using detectMultiscale() function.

5. Draw a rectangle around the detected face.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
import cv2

face_Cascade = cv2.CascadeClassifier("Resources/haarcascade_frontalface_default.xml")

image = cv2.imread('Resources/lena.png')
imgGray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

faces = face_Cascade.detectMultiScale(imgGray, 1.1, 4)

for (x, y, w, h) in faces:
  cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 0), 2)

cv2.imshow("Result", image)
cv2.waitKey(0)