Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| import requests | |
| import os | |
| from datasets import load_dataset, Image | |
| from PIL import Image | |
| from paddleocr import PaddleOCR | |
| from doctr.io import DocumentFile | |
| import torch | |
| # Set environment variable for PyTorch usage | |
| os.environ['USE_TF'] = '0' # Set TensorFlow to off | |
| os.environ['USE_TORCH'] = '1' # Set PyTorch to on | |
| from doctr.models import ocr_predictor | |
| device = torch.device("cuda" if torch.cuda.is_available() else "cpu") | |
| # Example for PyTorch model or doctr model | |
| ocr_model = ocr_predictor(det_arch='db_mobilenet_v3_large', reco_arch='crnn_mobilenet_v3_small', pretrained=True).to(device) | |
| import torch | |
| # Check if CUDA is available | |
| if torch.cuda.is_available(): | |
| print(f"GPU is available. Device: {torch.cuda.get_device_name(0)}") | |
| else: | |
| print("GPU is not available, using CPU.") | |
| """ | |
| Perform OCR with doctr | |
| """ | |
| def ocr_with_doctr(file): | |
| text_output = '' | |
| # Load the document | |
| doc = DocumentFile.from_pdf(file) | |
| # Perform OCR | |
| result = ocr_model(doc) | |
| # Extract text from OCR result | |
| for page in result.pages: | |
| for block in page.blocks: | |
| for line in block.lines: | |
| text_output += " ".join([word.value for word in line.words]) + "\n" | |
| return text_output | |
| """ | |
| Paddle OCR | |
| """ | |
| def ocr_with_paddle(img): | |
| finaltext = '' | |
| ocr = PaddleOCR(lang='en', use_angle_cls=True, use_gpu=True) | |
| # img_path = 'exp.jpeg' | |
| result = ocr.ocr(img) | |
| for i in range(len(result[0])): | |
| text = result[0][i][1][0] | |
| finaltext += ' '+ text | |
| return finaltext | |
| def generate_ocr(Method, file): | |
| text_output = '' | |
| if isinstance(file, bytes): # Handle file uploaded as bytes | |
| file = io.BytesIO(file) | |
| if file.name.endswith('.pdf'): | |
| # Perform OCR on the PDF using doctr | |
| text_output = ocr_with_doctr(file) | |
| else: | |
| # Handle image file | |
| img_np = np.array(Image.open(file)) | |
| text_output = generate_text_from_image(Method, img_np) | |
| return text_output | |
| def generate_text_from_image(Method, img): | |
| text_output = '' | |
| if Method == 'PaddleOCR': | |
| text_output = ocr_with_paddle(img) | |
| return text_output | |
| import gradio as gr | |
| image_or_pdf = gr.File(label="Upload an image or PDF") | |
| method = gr.Radio(["PaddleOCR"], value="PaddleOCR") | |
| output = gr.Textbox(label="Output") | |
| demo = gr.Interface( | |
| generate_ocr, | |
| [method, image_or_pdf], | |
| output, | |
| title="Optical Character Recognition", | |
| css=".gradio-container {background-color: lightgray} #radio_div {background-color: #FFD8B4; font-size: 40px;}", | |
| article="""<p style='text-align: center;'>Feel free to give us your thoughts on this demo and please contact us at | |
| <a href="mailto:[email protected]" target="_blank">[email protected]</a> | |
| <p style='text-align: center;'>Developed by: <a href="https://www.pragnakalp.com" target="_blank">Pragnakalp Techlabs</a></p>""" | |
| ) | |
| demo.launch(share=True) | |
| # import os | |
| # # Disable TensorFlow to ensure PyTorch is used | |
| # os.environ['USE_TF'] = '0' | |
| # import torch | |
| # print(torch.cuda.is_available()) # Should return True if GPU is available | |