ocr
**Document AI and OCR**
**Document Processing Pipeline**
```
[Document/Image]
|
v
[OCR: Image to Text]
|
v
[Layout Analysis]
|
v
[Structure Extraction]
|
v
[LLM Understanding]
```
**OCR Options**
| Tool | Strength | Use Case |
|------|----------|----------|
| Tesseract | Open source, good quality | General OCR |
| AWS Textract | Tables, forms | Enterprise docs |
| Google Doc AI | High accuracy, forms | Complex layouts |
| Azure Doc Intel | Structure extraction | Invoices, receipts |
| EasyOCR | Multilingual | Global documents |
**PDF Processing**
```python
# Extract text from PDF
from pypdf import PdfReader
def extract_pdf_text(path: str) -> str:
reader = PdfReader(path)
text = ""
for page in reader.pages:
text += page.extract_text()
return text
```
**Vision LLM for Documents**
Use multimodal LLMs to understand document images:
```python
def analyze_document_image(image_path: str, question: str) -> str:
return llm.generate_with_image(
image=image_path,
prompt=f"Analyze this document and answer: {question}"
)
```
**Table Extraction**
```python
def extract_tables(document: str) -> list:
return llm.generate(f"""
Extract all tables from this document as JSON arrays.
Each table should have headers and rows.
Document:
{document}
Tables (JSON):
""")
```
**Document Understanding Tasks**
| Task | Description |
|------|-------------|
| Classification | Categorize document type |
| Key-value extraction | Extract labeled fields |
| Table extraction | Parse tabular data |
| Question answering | Answer questions about doc |
| Summarization | Summarize document content |
**Chunking Strategies for PDFs**
```python
def chunk_pdf(pdf_path: str) -> list:
chunks = []
# By page
for page in extract_pages(pdf_path):
chunks.append({"type": "page", "content": page})
# By section (using headers)
sections = detect_sections(pdf_text)
for section in sections:
chunks.append({"type": "section", "title": section.title, "content": section.text})
return chunks
```
**Best Practices**
- Preprocess images (deskew, denoise) before OCR
- Combine OCR with layout analysis for tables
- Use multimodal LLMs for complex documents
- Validate extracted data against expected formats
- Handle multi-page documents appropriately