Fifteen free tools recover tables from scans, filings and charts — but no single one covers all three

Asked (summary):

Which free tools and methods recover tables from scanned PDFs, JavaScript-only filing viewers, and chart images — what does each tool miss, with one example document type per tool?

The shortlist covers 15 no-cost, open-source tools and methods — five per source type — researched on 10 Nov 2026 and backed by 12 independent source hosts, one source URL per row. These are representative free options, not every tool that exists. The pattern: scanned PDFs need OCR plus explicit table-structure recovery; JavaScript filing viewers are best bypassed through official machine-readable APIs or raw iXBRL plus an XBRL parser; chart images need axis calibration or chart-specific computer vision.

Scanned PDF or image JS-only filing viewer Chart image Bold outline = good default

Plain OCR recovers characters but drops rows and columns: Tesseract alone often loses column boundaries and headers even when the text itself is correct. unstract.com

The SEC Companyfacts API returns a whole company's tagged facts in one JSON call — but only standard-taxonomy facts; custom company tags never appear. sec.gov

WebPlotDigitizer is the verifiable manual route for charts; its one automation, AI Assist, is a closed-source cloud service outside the AGPL repository. hysenlabs.com

Automated chart models are draft extraction only: DePlot fails outright on pie charts and hallucinates heatmap values, so results need human checking. medium.com

All 15 tools: how each works, what it misses, one example document

Source type	Tool	How it recovers the table	What it misses	Example document	Free basis	Source

Method: 15 free or open-source table-recovery tools and methods (one row each), five per source type — scanned PDF, JavaScript-only filing viewer, chart image — researched 2026-11-10 across 12 independent source hosts, one distinct source URL per row. "What it misses" is the row source's own stated limitation, sometimes from third-party tests or project papers; method and limitation texts are condensed for space, with full wording at each linked source. Good defaults follow the brief's recommendations.

This report was generated automatically by Keenable SELECT at a user's request, from publicly available web sources linked herein. Keenable does not review, verify, or endorse its contents and makes no representation as to accuracy, completeness, or timeliness; AI-based extraction may contain errors. Nothing in this report is investment, legal, financial, or other professional advice. All trademarks and referenced content remain the property of their respective owners; no affiliation or endorsement is implied. To report an error, rights concern, or request removal: legal@keenable.ai.