Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesAn OCR job can finish, return a plausible PDF and report no error while producing no searchable words at all. In a September 21, 2026 first-person postmortem, author Jakub Wietrzyk describes exactly that failure on his site: “The text layer inside it was empty.” A reader might not notice until trying Ctrl+F. The lesson is to validate the words in the generated document, not just whether the job completed.
What the incident showed
Wietrzyk reports that his browser-based OCR workflow ran for months and returned PDFs that looked valid, even though their searchable text layer was empty. A known word list exposed the mismatch: for scan-150dpi-5p.pdf, the “Searchable PDF” output had 0.0% word recall, while “Text only” output had 100.0% recall. Those are the author’s results for that sample, not general OCR accuracy figures.
As an Amazon Associate I earn from qualifying purchases.
The case is a useful example of plausible success instead of an error. A file can be produced and still fail at several points between rendering a scanned page and writing extractable text into the final PDF.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Why OCR returned a file without useful text
A page-size threshold changed the rendering path
In Wietrzyk’s reported pdf.js path, ImageResizer reached the default DOMCanvasFactory only for pages exceeding a 2048-pixel dimension threshold. He gives a 300 dpi A4 page as 2481 × 3507 pixels, which took that path inside a worker where document was unavailable. A 150 dpi A4 page, at 1240 × 1754 pixels, stayed below the threshold. Thus a larger scan could fail while a smaller one appeared to work. His fix was to inject a worker-compatible canvas factory based on OffscreenCanvas. These details describe his implementation, not a guarantee about every pdf.js version or application. Read the case report’s rendering-path account.
#1 Best Overall
- 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
The code expected the wrong result structure
The code expected result.data.words, but the tesseract.js v7 words were nested under block, paragraph, line and word structures. An empty-array fallback turned an undefined field into what looked like a valid result containing zero words. After changing the code to use data.blocks, the author encountered another issue: blocks were disabled unless explicitly requested. A missing field or a null value is not evidence that the recognizer examined the page and found no text. Confirm the result shape for the library version in use, and request optional structured output explicitly. See the reported result-shape and configuration problems.
Font encoding dropped multilingual text
Wietrzyk says pdf-lib’s default WinAnsi font could not encode much of the language output his site offered. Exceptions thrown while writing words were swallowed by an empty catch, so text could disappear without a visible failure. He describes embedding a font and round-tripping generated PDFs to check whether text survived. Chinese, Japanese and Hindi failed that round trip and were removed from the offered set; Arabic passed the encoding round trip, but its recognition accuracy was not measured, so it remained excluded. These are implementation-specific findings, not a current compatibility guarantee for pdf-lib or a general statement about OCR quality in those languages. Read the author’s account of font encoding and language checks.
Rank #2
- 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)
An internal exception was reported as a bad input file
The post describes an error handler that searched for the substring “read.” It classified an internal “Cannot read properties of undefined” exception as a damaged input PDF and advised the user to rescan. That blamed a good document for a software failure. The case illustrates why error classification should distinguish faults in the input from exceptions in rendering, result parsing or PDF writing; the post does not specify a replacement implementation. See the report’s description of the misleading error.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →How to tell whether OCR output actually contains the right words
Check the generated artifact against known text, using the real route your users take. A completion status, a nonempty PDF, a plausible page count or a recognizer confidence value can miss failures downstream: rendering may take the wrong path, parsed results may be empty, or text may fail during PDF writing.
Rank #3
- 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
- Prepare ground truth. For each test scan, record the words that should be recognized. Use representative documents, including different page sizes, resolutions and every language the application claims to support.
- Exercise the end-to-end path. Run the actual site in a real browser, including rendering, OCR, result parsing and generated-PDF writing. Unit tests can check individual functions, but they do not establish that the complete browser workflow produces useful text.
- Extract or inspect the output text. Test the PDF that the application actually generated, not only an intermediate OCR response. For text-only output, inspect that output as well.
- Compare against the known words. Measure word recall—the share of ground-truth words present in the output—and review the missing words and document behavior. Keep the metric tied to its sample and conditions; one sample’s result is not an estimate for other documents or OCR systems.
- Test edge cases deliberately. Include pages on both sides of any size or resolution thresholds, and round-trip generated text for each supported language. Verify that absent optional result data becomes an explicit diagnostic rather than a silent empty-success result.
- Record the run’s origin. Preserve the environment or origin associated with every measurement. Wietrzyk says a late localhost run overwrote production results with pre-fix figures; his harness was changed to compare the recorded origin and abort before measuring. Mixing environments can make before-and-after numbers misleading.
Wietrzyk summarizes the purpose of this check: “Word recall against ground truth is a number that cannot be satisfied by code that merely finishes.” His reported production before-and-after measurements were:
| Sample PDF | Before | After |
|---|---|---|
scan-clean-300dpi-3p.pdf |
Page-one crash | 100.0% word recall; 2.0 seconds |
scan-150dpi-5p.pdf |
0.0% word recall | 100.0% word recall; 5.0 seconds |
scan-300dpi-10p.pdf |
Page-one crash | 100.0% word recall; 5.0 seconds |
These are results reported by Wietrzyk for his production workflow and named samples in 2026, not independently audited measurements or a benchmark of OCR systems generally. The post credits an end-to-end harness using the real site in a real browser and documents with known correct word lists. Read the reported validation method and measurements. The article also says the OCR engine and language data were fetched on first use, so the site’s first-use path did not work offline. Read the author’s account and its stated scope.
Rank #4
- Scanner type: Document
- Connectivity technology: USB
- With Auto Scan Mode, the scanner automatically detects what you're scanning
- Digitize documents and images
What the case does—and does not—establish
This is a detailed first-person report by Jakub Wietrzyk about his own codebase, published September 21, 2026. It documents how several failures combined to create a plausible success: a returned PDF and no clear error, but no usable text. It does not show how frequently these defects occur across OCR products, independently verify the author’s measurements, or establish current behavior across versions of tesseract.js, pdf.js, pdf-lib or browser worker APIs. Treat its version-specific findings as a diagnostic case study, not universal API guidance.
Free tools Windows power users keep installed
One-click scans. No signup required.
Quick Recap
Best Value
- OUR MOST ADVANCED SCANSNAP. Large touchscreen, fast 45ppm double-sided scanning, 100-sheet document feeder, Wi-Fi and USB connectivity, automatic optimizations, and support for cloud services. Upgraded replacement for the discontinued iX1600
- CUSTOMIZABLE. SHARABLE. Select personalized profiles from the touchscreen. Send to PC, Mac, mobile devices, and clouds. QUICK MENU lets you quickly scan-drag-drop to your favorite computer apps
- STABLE WIRELESS OR USB CONNECTION. Built-in Wi-Fi 6 for the fastest and most secure scanning. Connect to smart devices or cloud services without a computer. USB-C connection also available
- PHOTO AND DOCUMENT ORGANIZATION MADE EFFORTLESS. Easily manage, edit, and use scanned data from documents, receipts, photos, and business cards. Automatically optimize, name, and sort files
- AVOIDS PAPER JAMS AND DAMAGE. Features a brake roller system to feed paper smoothly, a multi-feed sensor that detects pages stuck together, and skew detection to prevent paper damage and data loss
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.
