Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| from PIL import Image | |
| import pytesseract | |
| import spacy | |
| from reportlab.lib.pagesizes import letter | |
| from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Image as RLImage | |
| from reportlab.lib.styles import getSampleStyleSheet | |
| import os | |
| from datetime import datetime | |
| # Load NLP model for analysis | |
| try: | |
| nlp = spacy.load("en_core_web_sm") | |
| except OSError: | |
| os.system("python -m spacy download en_core_web_sm") | |
| nlp = spacy.load("en_core_web_sm") | |
| def analyze_text(text): | |
| """Analyze the writing style and content""" | |
| doc = nlp(text) | |
| # Calculate word count and lengths | |
| words = [token for token in doc if not token.is_punct] | |
| word_count = len(words) | |
| word_lengths = sum(len(token) for token in words) | |
| analysis = { | |
| "word_count": word_count, | |
| "sentence_count": len(list(doc.sents)), | |
| "avg_word_length": word_lengths / max(1, word_count), | |
| "pos_tags": {tag: 0 for tag in ["NOUN", "VERB", "ADJ", "ADV"]}, | |
| "readability": "Moderate", | |
| "sentiment": "Neutral" | |
| } | |
| # Count POS tags | |
| for token in doc: | |
| if token.pos_ in analysis["pos_tags"]: | |
| analysis["pos_tags"][token.pos_] += 1 | |
| # Simple readability assessment | |
| if analysis["avg_word_length"] < 4: | |
| analysis["readability"] = "Easy" | |
| elif analysis["avg_word_length"] > 6: | |
| analysis["readability"] = "Difficult" | |
| return analysis | |
| def create_pdf(text, analysis, image_path=None): | |
| """Generate PDF report""" | |
| timestamp = datetime.now().strftime("%Y%m%d_%H%M%S") | |
| pdf_path = f"handwriting_report_{timestamp}.pdf" | |
| doc = SimpleDocTemplate(pdf_path, pagesize=letter) | |
| styles = getSampleStyleSheet() | |
| story = [] | |
| # Title | |
| story.append(Paragraph("Handwriting Analysis Report", styles["Title"])) | |
| story.append(Spacer(1, 12)) | |
| # Add original image if provided | |
| if image_path and os.path.exists(image_path): | |
| try: | |
| img = RLImage(image_path, width=400, height=300) | |
| story.append(img) | |
| story.append(Spacer(1, 12)) | |
| except: | |
| pass | |
| # Extracted Text section | |
| story.append(Paragraph("Extracted Text:", styles["Heading2"])) | |
| story.append(Spacer(1, 12)) | |
| story.append(Paragraph(text, styles["Normal"])) | |
| story.append(Spacer(1, 24)) | |
| # Analysis section | |
| story.append(Paragraph("Writing Analysis:", styles["Heading2"])) | |
| story.append(Spacer(1, 12)) | |
| analysis_text = f""" | |
| Word Count: {analysis['word_count']}<br/> | |
| Sentence Count: {analysis['sentence_count']}<br/> | |
| Average Word Length: {analysis['avg_word_length']:.2f} characters<br/> | |
| Readability: {analysis['readability']}<br/> | |
| <br/> | |
| Part-of-Speech Distribution:<br/> | |
| - Nouns: {analysis['pos_tags']['NOUN']}<br/> | |
| - Verbs: {analysis['pos_tags']['VERB']}<br/> | |
| - Adjectives: {analysis['pos_tags']['ADJ']}<br/> | |
| - Adverbs: {analysis['pos_tags']['ADV']}<br/> | |
| <br/> | |
| Overall Sentiment: {analysis['sentiment']} | |
| """ | |
| story.append(Paragraph(analysis_text, styles["Normal"])) | |
| doc.build(story) | |
| return pdf_path | |
| def process_handwriting(image): | |
| """Main processing function""" | |
| try: | |
| # Convert image to text | |
| text = pytesseract.image_to_string(image) | |
| if not text.strip(): | |
| return "No text could be extracted from the image.", None | |
| # Analyze the text | |
| analysis = analyze_text(text) | |
| # Save image temporarily for PDF | |
| temp_image_path = "temp_input_image.jpg" | |
| image.save(temp_image_path) | |
| # Create PDF report | |
| pdf_path = create_pdf(text, analysis, temp_image_path) | |
| # Clean up | |
| if os.path.exists(temp_image_path): | |
| os.remove(temp_image_path) | |
| return text, pdf_path | |
| except Exception as e: | |
| return f"Error processing image: {str(e)}", None | |
| # Gradio interface | |
| with gr.Blocks() as demo: | |
| gr.Markdown("# Handwriting Analysis Tool") | |
| gr.Markdown("Upload a handwritten image to extract text and get a writing analysis report.") | |
| with gr.Row(): | |
| with gr.Column(): | |
| image_input = gr.Image(label="Upload Handwritten Image", type="pil") | |
| submit_btn = gr.Button("Analyze") | |
| with gr.Column(): | |
| text_output = gr.Textbox(label="Extracted Text", lines=10) | |
| pdf_output = gr.File(label="Download Analysis Report") | |
| submit_btn.click( | |
| fn=process_handwriting, | |
| inputs=image_input, | |
| outputs=[text_output, pdf_output] | |
| ) | |
| if __name__ == "__main__": | |
| demo.launch() |