Update README.md

README.md

@@ -11,9 +11,119 @@ pinned: false
 short_description: A simpe app that counts characters including spaces in PDF
 ---
 
-# Welcome to Streamlit!
+# PDF Character Counter
 
-Edit `/src/streamlit_app.py` to customize this app to your heart's desire. :heart:
+A simple and reliable PDF character counting tool that extracts text from PDF documents and counts only the actual content.
 
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+Unlike many basic PDF counters, this tool automatically detects and removes:
+
+* Page numbers
+* Repeated headers
+* Repeated footers
+* Running chapter headers
+
+This makes it useful for academic papers, theses, reports, assignments, and other documents where administrative text should not be included in the final character count.
+
+## Features
+
+✅ Accurate PDF text extraction
+
+✅ Automatic page number removal
+
+✅ Automatic header detection
+
+✅ Automatic footer detection
+
+✅ Running chapter header detection
+
+✅ Per-page character statistics
+
+✅ Optional page exclusion
+
+✅ Detailed diagnostics showing removed content
+
+## How It Works
+
+The application processes PDFs in six steps:
+
+1. Extract all text blocks from every page.
+2. Detect recurring headers and footers based on position and repetition.
+3. Identify page numbers using pattern matching.
+4. Detect running chapter headers.
+5. Remove non-content elements.
+6. Count characters in the remaining text.
+
+Only the cleaned content is included in the final count.
+
+## What Gets Removed
+
+### Page Numbers
+
+Examples:
+
+```text
+Page 12
+12
+12 / 120
+12 of 120
+Side 12
+12 af 120
+```
+
+### Running Headers
+
+Examples:
+
+```text
+2.1 Methods 12
+4.3 Results iv
+```
+
+### Repeated Headers and Footers
+
+Text that appears on multiple pages in the top or bottom regions of the document is automatically detected and excluded.
+
+## Output
+
+The tool returns:
+
+* Total character count
+* Character count per page
+* Included text
+* Detected headers
+* Detected footers
+* Detected running headers
+* Detected page numbers
+* Log of removed elements
+
+## Use Cases
+
+* Academic assignments
+* University theses
+* Research papers
+* Government reports
+* Technical documentation
+* Publication word/character limit verification
+
+## Technology
+
+Built with:
+
+* Python
+* PyMuPDF (fitz)
+* Regular Expressions
+* Hugging Face Spaces
+
+## Limitations
+
+* Scanned PDFs without embedded text require OCR before processing.
+* Very unusual document layouts may affect automatic header/footer detection.
+* Character counts are based on extracted text and may differ slightly from counts generated by word processors.
+
+## License
+
+MIT License
+
+## Author
+
+Created by Daniel Hjerresen.