camelot-dev/camelot
Camelot
Camelot is a Python library and processing engine designed to extract tabular data from PDF documents. It converts unstructured tables into machine-readable formats such as CSV, JSON, and Excel.
The project provides specialized toolsets for different document types, using line detection for ruled tables and whitespace analysis for borderless tables. It includes an optical character recognition system to recover structured data from image-based scanned PDFs that lack a digital text layer.
The library handles complex document layouts, including encrypted files, rotated pages, and tables that span multiple pages. It supports parallel execution across multiple CPU cores to increase processing speed and includes validation systems to calculate confidence scores for extraction accuracy.
A command line interface is available for executing extraction tasks directly from the terminal.
Features
- Table Extraction - Provides a specialized engine for extracting tabular data from PDF documents into machine-readable formats.
- PDF Table Parsing - Provides a system for parsing tabular data from documents using line or whitespace detection and rotated page handling.
- OCR Document Parsers - Uses optical character recognition to extract structured information from scanned image-based PDFs.
- OCR Tools - Combines optical character recognition with structure detection to extract data from scanned PDFs.
- OCR Layout Integration Pipelines - Integrates layout analysis with optical character recognition to recover structured data from image-based PDFs.
- Table Cell Detections - Implements lattice-based line detection to identify individual table cells within PDF document structures.
- Table Extraction - Extracts structured tabular data from PDF documents and converts it into machine-readable formats like CSV or Excel.
- PDF to CSV Converters - Transforms unstructured PDF table layouts into structured CSV, JSON, or Excel files for data analysis.
- Heuristic Table Identification - Uses heuristic methods to identify and parse tabular data from borderless or ruled PDF documents.
- PDF Parsers - Provides a processing engine to parse and convert content from PDF documents, including encrypted and rotated pages.
- Visual Table Extraction - Integrates structure detection and OCR to convert image-based tabular data into structured digital text.
- Whitespace-Based Table Analysis - Extracts tabular data from borderless documents by analyzing the spacing between text characters and words.
- Automated Data Entry Tools - Extracts structured data from document tables to automate the population of databases and avoid manual entry.
- Document Digitization Tools - Handles complex PDF layouts, including encrypted files and multi-page tables, to digitize business data.
- Concurrent Data Processing - Implements parallel execution across multiple CPU cores to increase the speed of PDF page parsing.
- Cross-Page Table Reconciliation - Stitches fragmented table segments across page breaks into a single unified data structure using coordinate matching.
- PDF Extraction Pipelines - Builds automated workflows that parse tables from multiple PDF pages and validate extraction accuracy for production.
- Document Table Merging - Combines table continuations spanning across multiple pages into a single unified data structure.
- Multi-Format Data Exports - Converts extracted tabular data into common file types such as CSV, JSON, and Excel for cross-tool compatibility.
- Document Processing Parallelization - Distributes PDF page parsing across multiple CPU cores to accelerate the extraction of large documents.
- Data Loading Extraction - Extract tabular data from PDFs.
- Document Parsing - Extracts structured tabular data from PDFs into DataFrames.
- Documentation and Processing - Library for extracting tables from PDFs.
