I have a PDF document that I want to scan with pytesseract, but the page numbers are not recognized. The page number is not recognized on any of the pages. The PDF is written with Latex. I ried different psm, but it did not help. What can I do that tesseract recognizes the page numbers? The PDF document gets uploaded via streamlit and is passed to the function as a BytesIO like object. The function returns an array of words (strings). Code: def get_text_from_ocr(uploaded_file): images = [] config = r"--psm 3" # 3: Fully automatic page segmentation, but no OSD. (Default) # pdf to images uploaded_file.seek(0) pdf_bytes = uploaded_file.read() doc = pymupdf.open(stream=pdf_bytes, filetype="pdf") for page in doc: pix = page.get_pixmap(dpi=300) img = Image.open(BytesIO(pix.tobytes("png"))) images.append(img) # Do OCR text = [word for img in images for word in pytesseract.image_to_string(img, config=config).split()] return text I also tried some preprocessing, but it did not help either (convert to binary and enlarge the image). Code: def get_text_from_ocr(uploaded_file): images = [] config = r"--psm 3" # 3: Fully automatic page segmentation, but no OSD. (Default) # pdf to images uploaded_file.seek(0) pdf_bytes = uploaded_file.read() doc = pymupdf.open(stream=pdf_bytes, filetype="pdf") for page in doc: pix = page.get_pixmap(dpi=300) img = Image.open(BytesIO(pix.tobytes("png"))) # Preprocessing gray = img.convert("L") # "L" = 8-bit grayscale # Tune threshold value as needed (e.g., 180, 200) binary = gray.point(lambda x: 0 if x < 180 else 255, '1') # '1' mode = black & white scale = 2 resized = img.resize((img.width * scale, img.height * scale), Image.LANCZOS) images.append(resized) # Do OCR text = [word for img in images for word in pytesseract.image_to_string(img, config=config).split()] return text PDF: PDF document Page 1 as image: Page 2 as image: Continue reading...
