Alphabet classification Using Computer Vision

user_14441659 0
Viewed 92

I am a beginner to image processing. I am trying to detect A,B,C,D,1,2 Using computer vision. How can I detect & classify this image without deep learning. The images are A, B, C, 2

My assumption if it can be solved by contour analysis. But have no idea how to do. Any implementation idea or help will be highly appreciated. Thanks in advance.

computer-vision opencv python
1 Answers

As all your letters and numbers always appear to completely white, I'd preprocess your images first with some simple thresholding at high intensity grayscale values:

src = cv.imread("A.png", 0)
ret,thresh = cv.threshold(src, 245, 255, cv.THRESH_BINARY)

You will get something like this as a result:

enter image description here

After this there is a number of ways you could detect which number/letter is present on the image. You could use a available OCR package like pytesseract or manually implement template matching for every available number/letter.

user_64369770
Related