3 Online OCR Services to Convert Scanned Docs to Text

Understanding Optical Character Recognition

OCR, which stands for Optical Character Recognition, is a technology with origins dating back to the late 1920s. The first patent for OCR was issued in Germany during this period.

Currently, OCR is primarily employed to transform scanned documents into editable text formats. This capability proves invaluable when extracting text from images or when needing to modify a scanned document’s content.

This article details three freely accessible online OCR services, evaluated for their feature sets and overall quality of text conversion.

OnlineOCR

This particular service excels at recognizing both text and characters from both PDF and image files. Converted files can be output in various formats, including PDF and Microsoft Word.

A significant advantage is its ability to preserve original formatting elements like tables, columns, bullets, and graphics – a feature not commonly found in other web-based OCR tools.

The service is usable without registration, though limitations apply. Uploaded files are restricted to a maximum size of 1MB. The default language setting is English and cannot be altered, and the output is provided as plain text without formatting.

Creating an account unlocks additional features. Users gain access to multiple language options, a wider range of output formats, and the ability to upload several files simultaneously, up to a combined total of 30MB. New users receive 5 credits, with each credit allowing the processing of a single or multi-page image file.

A Demonstration mode is available if you prefer not to use a credit. This allows you to preview the results and assess the expected quality. Additional credits can be purchased, starting at a cost of 6 cents per page.

During testing, I utilized a scanned formal letter containing visible folding edges and shading. Despite these imperfections, the service accurately recognized both the text and the original formatting.

However, it’s important to note that this service does not support the recognition of handwritten text.

NewOCR

Free Online OCR offers a simple and direct approach to text extraction. There are no limits on the number of files you can upload.

File size restrictions are in place: image files must be under 5MB, and PDF documents should not exceed 20MB. Users can select a language during upload to optimize recognition accuracy and rotate the document as needed. Registration is not required for use.

Unfortunately, all formatting is lost during conversion, and handwritten text is not recognized. Furthermore, in my testing, portions of the scanned document appeared in the header. Therefore, this tool is best suited for quickly obtaining plain text for copying and pasting into another document.

Free OCR

Free OCR accepts image and PDF files up to 2MB in size. It will only recognize the first page of a PDF document.

Selecting the document’s language can improve the recognition of language-specific characters. A notable feature is the ability to upload up to 10 images per hour. While multi-column text is supported, the output is always plain text, meaning formatting is not preserved. There is no membership or subscription option.

Despite its popularity, this service yielded the poorest results in my tests. Characters accurately recognized by other services were often missed by Free OCR. The presence of folding edges in the scan image appeared to be a contributing factor.

Additionally, sections of the text were displayed in a random order, potentially due to the service’s method of handling formatted text.

Karl has authored a comprehensive article detailing Free OCR. You can find additional tips and techniques for achieving better results here: Convert Images To Text Online With Free OCR

If you are interested in exploring further web-based OCR options, WeOCR is worth investigating.

As demonstrated, utilizing a web-based OCR service is generally suitable for quick and basic text recognition tasks. For scenarios requiring the scanning of numerous documents with a need for perfectly formatted output, I recommend downloading and installing dedicated OCR software.