Spaces:
Running
Running
| from transformers import TrOCRProcessor, VisionEncoderDecoderModel | |
| from PIL import Image, ImageFilter | |
| import torch | |
| import re | |
| # Load TrOCR model and processor | |
| processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-handwritten") | |
| model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-handwritten") | |
| def clean_ocr_text(text): | |
| text = text.replace(",", ".").replace("s", "5").replace("o", "0").replace("O", "0") | |
| return re.sub(r"[^\d.kg]", "", text.lower()) | |
| def restore_decimal(text): | |
| if re.fullmatch(r"\d{5}", text): | |
| return f"{text[:2]}.{text[2:]}" | |
| elif re.fullmatch(r"\d{4}", text): | |
| return f"{text[:2]}.{text[2:]}" | |
| return text | |
| def extract_unit_from_text(raw_text): | |
| raw_text = raw_text.lower() | |
| if "kg" in raw_text or "kgs" in raw_text or "k9" in raw_text: | |
| return "kg" | |
| elif re.search(r'\dg', raw_text) or "gram" in raw_text: | |
| return "g" | |
| return None | |
| def extract_weight(image): | |
| try: | |
| image = image.resize((image.width * 2, image.height * 2), Image.BICUBIC) | |
| image = image.filter(ImageFilter.SHARPEN) | |
| pixel_values = processor(images=image, return_tensors="pt").pixel_values | |
| generated_ids = model.generate(pixel_values) | |
| raw_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0].strip() | |
| cleaned = clean_ocr_text(raw_text) | |
| match = re.search(r"(\d{1,3}\.\d{1,3})\s*(kg|g)?", cleaned) | |
| if match: | |
| weight = match.group(1) | |
| unit = match.group(2) or extract_unit_from_text(raw_text) | |
| return f"{weight} {unit or ''}".strip(), raw_text | |
| match = re.search(r"\d{4,5}", cleaned) | |
| if match: | |
| decimal_fixed = restore_decimal(match.group()) | |
| unit = extract_unit_from_text(raw_text) | |
| return f"{decimal_fixed} {unit or ''}".strip(), raw_text | |
| match = re.search(r"\d+", cleaned) | |
| if match: | |
| unit = extract_unit_from_text(raw_text) | |
| return f"{match.group()} {unit or ''}".strip(), raw_text | |
| return "Error: No valid weight found", raw_text | |
| except Exception as e: | |
| return f"Error: {str(e)}", "" | |