I'm attempting to perform OCR on a set of single letters inside an image using Python. I'm new to this so apologies if I get the terminology wrong, but I've filtered and have obtained (I think) quite plain images of my regions of interest. I'm then using the following to get the character from my images. config = "--psm 10 --oem 3 -c tessedit_char_whitelist=ABCDEFGHIJKLMNOo0PQRSTUVWXYZ" text = pytesseract.image_to_string(roi_resized, config=config) This works for all letters except the letter O. Everything else appears fine but O doesn't detect. The exact image I'm using is shown below. Is there anything I can try? I've already tried increasing the padding around the image to no avail. I've also tried putting other characters in the whitelist such as lower case o and the number 0. If this won't work, perhaps because of the font(?), can anyone recommend any alternative solutions please? Thanks Continue reading...
