Free tools Windows power users keep installed
One-click scans. No signup required.
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
To extract a table from a PDF, first check whether its text is selectable, then choose a method that matches the table’s layout: line detection for ruled tables, text alignment or whitespace for borderless tables, and OCR plus table-structure recognition for scans. Extraction is not simply copying a table stored inside the file. Many PDFs position text and draw lines without recording which words belong to which rows, columns, or cells, so the result must be checked against the page.
Why PDF table extraction is a layout problem
A PDF can show a clear grid to a reader while storing only positioned characters and drawing operations. It may not contain explicit row, column, or cell relationships. An extractor therefore has to locate the table, estimate its boundaries, infer rows and columns, and assign text to cells by position. The task becomes less predictable with borderless layouts, irregular spacing, merged cells, faint rules, or pages whose layouts vary.
This does not mean every PDF lacks structure. It means the file’s visible appearance does not guarantee that a table is represented as structured data. Tools such as Camelot and pdfplumber document approaches that infer tables from lines, whitespace, and text alignment.
Choose a method based on the PDF and table
Text-based PDFs with visible rules
Start with a line-oriented parser when the table has clear horizontal and vertical rules. Camelot’s Lattice parser detects rules and their intersections, then uses those boundaries to assign text to cells. Faint or incomplete rules can make detection harder; Camelot’s current documentation also describes combined vector and raster detection.
#1 Best Overall
- Scanner type: Document
- Connectivity technology: USB
- With Auto Scan Mode, the scanner automatically detects what you're scanning
- Digitize documents and images
Text-based PDFs without grid lines
For columns separated mainly by whitespace, try a method that uses text positions and alignment. Camelot’s Stream parser groups text into rows and estimates columns from whitespace and word alignment. Its Network parser looks for repeated text-box alignments, while Hybrid combines text alignment with line-derived boundaries. These are different ways to infer geometry, not guarantees that every layout will be read correctly.
Scanned or image-only PDFs
If the table is only an image, a text-based parser has no selectable words to place into cells. Use an OCR and table-structure-recognition path instead. Camelot’s 2.0.0 documentation describes an optional machine-learning route, flavor='ml', that uses Table Transformer and can support image-only PDFs when OCR is configured. The project notes that this route adds dependencies, including PyTorch and OCR components as needed, and runs more slowly than its heuristic parsers. Check your installed Camelot version and the optional extras before following those instructions: older FAQ wording that says Camelot only works with text-based PDFs does not reflect the current 2.0.0 overview. Camelot’s installation and optional-dependency guidance describes setup.
Rank #2
- PORTABLE SCANNER FOR USE ON-THE-GO — The fastest and lightest mobile single-sheet-fed compact document scanner in its class¹
- QUICK DOCUMENT SCANNING ― This Epson ultra-fast scanner scans a single page as quickly as 5.5 seconds²; Windows and Mac compatible
- VERSATILE PAPER HANDLING ― Portable scanner scans documents up to 8.5 x 72 in; Also easily digitizes receipts and ID cards to make accounting, bookkeeping, and organizing simpler
- INTUITIVE, HIGH-SPEED SOFTWARE — Epson ScanSmart Software³ is a smart tool allowing you to easily scan, review, and save; Stay organized easily with the help of this Epson scanner
- EASY SETUP — USB-powered connect to your computer for quick and simple scanning; No batteries or external power supply required to operate portable document scanner; Standard Connectivity: USB 2.0
When you need to inspect or tune PDF objects
pdfplumber is useful for examining PDF objects and tuning table detection on machine-generated PDFs. It can find explicit lines or infer them from aligned text, and it provides table-debugging information and page cropping. It does not provide OCR and says it lacks strong support for extracting tables from OCRed documents, so it is not the right standalone path for a scanned page.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →When a hosted service fits the workload
A hosted document-analysis service such as Amazon Textract can return table and cell relationships, including information such as headers and merged cells. AWS also documents asynchronous processing for multipage PDF workloads. This route requires cloud integration and service setup; it is an option for a workflow that needs a hosted API, not a universal accuracy winner.
Rank #3
- FAST SPEEDS - Scans color and black and white documents a blazing speed up to 16ppm (1). Color scanning won’t slow you down as the color scan speed is the same as the black and white scan speed.
- ULTRA COMPACT – At less than 1 foot in length and only about 1. 5lbs in weight you can fit this device virtually anywhere (a bag, a purse, even a pocket).
- READY WHENEVER YOU ARE – The DS-640 mobile scanner is powered via an included micro USB 3. 0 cable allowing you to use it even where there is no outlet available. Plug it into you PC or laptop and you are ready to scan.
- WORKS YOUR WAY – Use the Brother free iPrint&Scan desktop app for scanning to multiple “Scan-to” destinations like PC, Network, cloud services, Email and OCR. (2) Supports Windows, Mac and Linux and TWAIN/WIA for PC/ICA for Mac/SANE drivers. (3)
- OPTIMIZE IMAGES AND TEXT – Automatic color detection/adjustment, image rotation (PC only), bleed through prevention/background removal, text enhancement, color drop to enhance scans. Software suite includes document management and OCR software. (4)
A practical extraction workflow
- Check the text layer. Open the PDF and try selecting and copying the table text. Selectable words indicate that a text-based approach may work. If the page behaves like a single image, plan for OCR and structure recognition rather than assuming a normal text parser can recover the cells.
- Classify the layout. Note whether rules are visible, columns are separated by whitespace, more than one table appears on the page, headers are merged, text is rotated, or the table continues onto another page. Those features affect table boundaries and cell assignment.
- Start with the matching parser. Try a line-based method for a ruled text table and an alignment or whitespace method for a borderless text table. For scans, choose a tool path that explicitly supports OCR and table-structure recognition. With Camelot 2.0.0, the optional
flavor='ml'route is one documented choice for dense borderless tables and scanned documents when OCR is configured. - Inspect the detected region. Compare the extracted table area with the rendered page. If detection includes nearby text or misses part of the table, refine the page crop or table region and adjust the relevant settings. pdfplumber provides crop and table-debugging features; Camelot provides configurable extraction and filtering.
- Validate the extracted cells. Check the headers, first and last rows, column alignment, merged cells, and representative values against the page. Reconcile totals or row counts when the document supplies them. For consequential work, use confidence information where available and set human review according to the cost of an error.
What to expect from common approaches
| Approach | Good starting point | Tradeoffs to account for |
|---|---|---|
| Camelot Stream | Text-based tables with whitespace-separated columns | Infers geometry from text rows and alignment; settings may need tuning, and the result needs inspection. |
| Camelot Lattice | Text-based tables with visible rules | Uses rules and intersections; faint or incomplete lines can challenge detection. |
| Camelot Network or Hybrid | Tables with repeated text alignment, including some mixed layouts | Depends on text boxes and layout patterns; results vary with the document. |
| Camelot ML (2.0.0 documentation) | Dense borderless tables and scans when OCR is configured | Optional neural backend and additional dependencies; the project reports slower runtime than heuristic parsers. |
| pdfplumber | Inspecting PDF objects and tuning detection on machine-generated PDFs | Does not provide OCR and lacks strong support for tables in OCRed documents. |
| Amazon Textract | Hosted table analysis, including multipage PDF workloads | Requires cloud integration and service setup; merged or irregular cells can produce inconsistent results, and confidence scores need judgment. |
These approaches solve different input and workflow problems. Compare whether the file is text-based or scanned, how many pages and documents you need to process, what dependencies or cloud integration are acceptable, how sensitive the data is, and how much review the output requires. The available documentation does not establish a universal performance ranking or a direct comparison of cost, privacy, service levels, or accuracy across these options.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How much confidence should you put in benchmark figures?
Camelot’s project documentation reports results for its own benchmark on 545 financial PDFs in FinTabNet.c: F1 scores of 0.750 for ML, 0.725 for Network, and 0.658 for Hybrid. The same page reports TEDS values of 0.371, 0.200, and 0.198, respectively, but describes its TEDS measure as a cell-text proxy rather than exact tree-edit distance. These are project-published figures for that dataset, not independent comparisons or promises of accuracy on other kinds of PDFs. Camelot’s comparison documentation explains the figures and benchmark context.
Quick Recap
Best Value
- FITS SMALL SPACES AND STAYS OUT OF THE WAY. Innovative space-saving design to free up desk space, even when it's being used
- SCAN DOCUMENTS, PHOTOS, CARDS, AND MORE. Handles most document types, including thick items and plastic cards. Exclusive QUICK MENU lets you quickly scan-drag-drop to your favorite computer apps
- GREAT IMAGES EVERY TIME, NO EXPERIENCE REQUIRED. A single touch starts fast, up to 30ppm duplex scanning with automatic de-skew, color optimization, and blank page removal for outstanding results without driver setup
- SCAN WHERE YOU WANT, WHEN YOU WANT. Connect with USB or Wi-Fi. Send to Mac, PC, mobile devices, and cloud services. Scan to Chromebook using the mobile app. Can be used without a computer
- PHOTO AND DOCUMENT ORGANIZATION MADE EFFORTLESS. ScanSnap Home all-in-one software brings together all your favorite functions. Easily manage, edit, and use scanned data from documents, receipts, business cards, photos, and more
Rank #4
- FAST DOCUMENT SCANNING — Document scanner with feeder allows you to speed through stacks with a 50-sheet Auto Document Feeder (ADF); Efficient office scanner to help you scan more productively
- INTUITIVE, HIGH-SPEED SOFTWARE — Quickly scan with this desktop document scanner; Epson ScanSmart Software lets you easily preview scans, email files, upload to the cloud, and more; Plus, automatic file naming saves even more time
- SEAMLESS INTEGRATION — Easily incorporate your data into most document management software with the included TWAIN driver; Office document scanner integrates seamlessly with business workflows
- EASY SHARING — Duplex scanner allows you to scan straight to email or popular cloud storage2 services like Dropbox, Evernote, Google Drive, and OneDrive for simple storage and sharing
- SIMPLE FILE MANAGEMENT — Scanner allows the creation of searchable PDFs with Optical Character Recognition (OCR) and convert scans to editable Word or Excel files effortlessly; Designed for home and office document scanning
Common failure modes and how to respond
- The output contains no text or almost no cells: Check whether the page is image-only. If it is, switch to a supported OCR and table-recognition workflow.
- Nearby text is included as part of the table: Refine the crop or table region, then inspect the detected boundary again.
- Columns shift or values land in the wrong cells: Examine the text alignment and rules on the rendered page. Try the parser that better matches the layout, adjust its settings, and compare affected rows cell by cell.
- Merged or irregular cells are inconsistent: Treat the affected structure as a review point rather than assuming the parser has represented it correctly. Compare it with the page and check any available confidence information.
- Results vary across pages: Check whether the table layout changes, rotates, or continues across pages. A single set of boundaries may not fit every page; inspect and tune the affected regions individually.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
