pythonopencvtextocrmonospace

How can I insert Monospace fonts into an image with opencv?


Currently, I am able to insert some texts of HERSHEY font into images with openCV API (putText). But it seems openCV are not supporting any monospace font.

I was wondering how I can insert some Monospace or fixed-pitch texts into the image.


Solution

  • You could use PIL/Pillow for that aspect quite easily. OpenCV images are numpy arrays, so you can make a Pillow Image from an OpenCV image with:

    PilImage = Image.fromarray(OpenCVimage)
    

    Then you can draw with a mono spaced font using code in my answer here. You only need the 3 lines after the comment "Get a drawing context".

    Then you can convert back to OpenCV image with:

    OpenCVimage = np.array(PilImage)
    

    Note that you are not limited to monospaced fonts, you can use any Truetype font you like.

    That might look like this:

    #!/usr/local/bin/python3
    
    from PIL import Image, ImageFont, ImageDraw
    import numpy as np
    import cv2
    
    # Open image with OpenCV
    im_o = cv2.imread('start.png')
    
    # Make into PIL Image
    im_p = Image.fromarray(im_o)
    
    # Get a drawing context
    draw = ImageDraw.Draw(im_p)
    monospace = ImageFont.truetype("/Library/Fonts/Andale Mono.ttf",32)
    draw.text((40, 80),"Hopefully monospaced",(255,255,255),font=monospace)
    
    # Convert back to OpenCV image and save
    result_o = np.array(im_p)
    cv2.imwrite('result.png', result_o)
    

    enter image description here


    Alternatively, you could have a function generate a lump of canvas itself, write your text on it, and then splice it into your OpenCV image wherever you want. Something along these lines - though I have no idea of what flexibility you would require so I have not parameterised everything:

    #!/usr/local/bin/python3
    
    from PIL import Image, ImageFont, ImageDraw, ImageColor
    import numpy as np
    import cv2
    
    
    def GenerateText(size, fontsize, bg, fg, text):
       """Generate a piece of canvas and draw text on it"""
       canvas = Image.new('RGB', size, bg)
    
       # Get a drawing context
       draw = ImageDraw.Draw(canvas)
       monospace = ImageFont.truetype("/Library/Fonts/Andale Mono.ttf",fontsize)
       draw.text((10, 10), text, fg, font=monospace)
    
       # Change to BGR order for OpenCV's peculiarities
       return cv2.cvtColor(np.array(canvas), cv2.COLOR_RGB2BGR)
    
    
    # Open image with OpenCV
    im_o = cv2.imread('start.png')
    
    
    # Try some tests
    w,h = 350,50
    a,b = 20, 80
    text = GenerateText((w,h), 32, 'black', 'magenta', "Magenta on black")
    im_o[a:a+h, b:b+w] = text
    
    
    w,h = 200,40
    a,b = 120, 280
    text = GenerateText((w,h), 18, 'cyan', 'blue', "Blue on cyan")
    im_o[a:a+h, b:b+w] = text
    
    cv2.imwrite('result.png', im_o)
    

    enter image description here

    Keywords: OpenCV, Python, Numpy, PIL, Pillow, image, image processing, monospace, font, fonts, fixed, fixed width, courier, HERSHEY.