Skip to content

Latest commit

 

History

History
164 lines (76 loc) · 5.61 KB

Kavya_OpenCV_Basics.md

File metadata and controls

164 lines (76 loc) · 5.61 KB
Raw

OPENCV BASICS

  • What is OpenCV?

    OpenCV(*Open Source Computer Vision and Machine Learning Software)is an open source library which processes images and videos to identify objects, faces, detecting colours etc.It has wide applications in *Computer Vision,Machine Learning and *Image Processing. It also supports variety of programming languages like **Python, **C++, **C. Basically it a library for processing *images.

    OpenCV in Artificial Intelligence

    Computer Vision is a branch of Artificial Intelligence which trains the computer to extract information from digital data like images and videos, understand them and even communicate.

    Why OpenCV?

    Nowadays image processing and computer vision have gained importance in every field. As Opencv has over 2500 optimized algorithms which ease image processsing and even helps in building projects like tracking movements, recognizing faces, finding similar images etc.,it makes programming easier.Reading and displaying images is simplified and through OpenCV development of programs from simpler to complex is easier

    Application of OpenCV

    • Image enchancement.

    • Rotating,Cropping,Resizing(using more advanced features)

    • Background removal.

  • Images

    Image is collection of pixels and it is a binary representation of visual information such as logos, drawing pictures, graphs etc.

  • Black and white image

    The image which has only two colours i.e Black *and *white is called Binary image. For a basic black and white image there is only one bit representation where 0 represents *black and *1 represents white.

    image

  • Gray Scale Image

    One can have images of more than two levels i.e instead of having only 0 and 1 bit levels each pixel can have range of values i.e 2^8,this will give us resolution of 256 levels where 0 will be black and 255 will be *white. So basically we have 254 colours between *black and white.

    image

  • Coloured image

    For coloured images each pixel can have levels of red,green,blue. Different levels of red, green and blue give different colours to the respective pixels giving out a fully coloured image.

    image

  • Installation of OpenCV

    Installation of OpenCV has two steps to be followed through Anaconda Prompt.

    • Open Anaconda Prompt

    • Execute the following commands:

      pip install opencv-python

pip install opencv-contrib-python

After installing the Opencv package on anaconda prompt, for further usage of OpenCV in image and video processing through python IDE, it is necessary to import OpenCV library and it's functions using import cv2 statement

  • Basic functions in OpenCV

    • imread(): In order to read or store image in a variable imread() function is used.

      A variable is initialized to read an image, using the function imread() in cv2 package we store the image.

      syntax:

        variable_name = cv2.imread(specify_the_path_in_which_image_should_be_read_with_extensions)

      Example:

         img1 = cv2.imread("extras/nature.jpeg")

    • imshow(): This function is used to display the image from the variable where the image is stored. Window name represents the name of the window on which the image is to be displayed.

      syntax:

          cv2.imshow(window_name,variable_name)

      Example:

          cv2.imshow("Output",img1)

    • videocapture(): This function is used to import a video.

        - steps to import a video
  
     - Initialize a variable to store the imported video.
     
     - Using an infinite while loop and read() function, read the frames.
     
     - Use imshow() function to display the video.

      Example:

       cap=cv2.videocapture("specify_the_path_in_which_video_should_be_read_with_extensions")
 
  while true:
  
      success,img=cap.read()
      
      cv2.imshow("video",img)
      
      if cv2.waitKey(1)& 0xFF=ord('q'):
      
          break:

    • Displaying multiple images

      It is possible to display multiple images in a single window. It can be displayed either horizontly or vertically. One has to import numpy library for displaying multiple images.

      import numpy as np

      steps to display multiple images

      • store the multiple images in different variables using imread() function.
      • concatenate image Horizontally

      syntax:

      variablename=np.concatenate((image1,image2),axis=1)

      • concatenate image Vertically

      syntax:

      variable_name=np.concatenate((image1,image2),axis=0)

      Here axis refers to mode of concatenation.

      axis=1 refers to horizontal concatenation.

      axis=0 refers to horizontal concatenation.

      • Display the concatenated images using imshow() function.

    Contact me directly on my mail