AR.source.ocr namespace

AR.source.ocr.LabelProcessing module

AR.source.ocr.LabelProcessing.ReaderCroppedAndFullImage(img, regexp, type, background, range, ocrResults): Call ogLblShapeDetector to crop the image, call ReaderOneImageRecursive on the cropped image and on the full image if not succesfull to read the label. Returns a list containing all characters in the image satisfying the different part ot the regexp and a list containing all characters that were read on the image to be used if a new regexp is used on the same image.

AR.source.ocr.LabelProcessing.ReaderOneImageRecursive(img, regexp, type, ocrResults): Handles regexp by calling ogLblUtils, call ogLblOCR to perform OCR on the image provided

AR.source.ocr.OCR.PerformOCR(img): Use easyOcr module to perform OCR on the provided image. Returns a tuple containing the characters read and their position.

AR.source.ocr.OCR.ReplaceSymbol(text): Replace the ‘Ø’ charachter with an ‘0’

AR.source.ocr.OCR.ReturnMatch(text, labelMatcher): Match a text with a regexp. Returns a string containing the match. If none returns None.

AR.source.ocr.OCR.cleanup_text(text): strip out non-ASCII text so we can draw the text on the image using OpenCV

AR.source.ocr.ShapeDetector.HexaToRGB(hexa): Take a hexa color code and convert it to RGB. Returns a tuple with the RGB code.

AR.source.ocr.ShapeDetector.ShapeAndColorDetector(img, background, range): Perform HSV mapping on the provided image using a hexa color code or a predefined color. Crop the image according to this mask. Returns the new image.