OmniPage Ultimate is a Windows application for converting scanned paper, document photographs and PDFs into editable text, searchable files and form data. It combines optical character recognition with page correction tools and automated processing of incoming documents.
OmniPage Ultimate is paid software with a free trial. The Ultimate edition includes form creation, form data extraction, watched-folder and mailbox jobs, barcode-driven workflows and advanced PDF output controls. Network editions use separate licensing agreements.
Preparing pages for recognition
Pages can come directly from a scanner or from files such as TIFF, JPEG, PNG, BMP and PDF. Manual processing pauses between acquiring pages, recognizing them and exporting results, allowing different settings for individual pages.
Image tools can straighten a crooked scan using an angle or alignment line. For a photographed page, an adjustable grid corrects perspective. Cropping and rotation affect both the original image and the black-and-white copy used for recognition.
Brightness and contrast sliders adjust the original image, while despeckling removes stray dots from the recognition copy. Colour dropout removes a selected colour from that copy. Border and punch-hole tools remove marks around the page edges.
OmniPage divides pages into text, table and graphic zones. Users can draw these regions manually or let the program detect them. An ignore zone excludes its contents from both recognized text and transferred images.
Recognizing and correcting text
OmniPage Ultimate recognizes more than 120 languages. Selecting the document’s languages determines which characters the recognition process considers. Dictionary support covers a smaller group of languages and supplies word suggestions during correction.
The Proofreader presents a suspect word in its surrounding text beside the original page image. Users can type a correction, choose a suggestion or leave the word unchanged. Change All applies a correction to repeated occurrences.
Adding an accepted word to a user dictionary makes it available for later recognition and proofreading. Proofing marks remain in the OmniPage workfile but disappear from exported text. Chinese, Japanese and Korean recognition excludes dictionaries, proofreading and training.
Chinese, Japanese and Korean must also be processed separately from other recognition languages. For other supported combinations, selecting several languages makes their characters available together; languages with dictionaries can contribute suggestions during the same proofing session.
Exporting documents and tables
Recognized pages can be exported to Word, Excel, RTF, plain text, CSV, WordPerfect, HTML or ePub, among other formats. Users choose the current page, selected pages or the complete document, then select the available layout treatment.
| Layout choice | Result |
|---|---|
| Plain Text | A single stream of text with one font and size. TXT output omits graphics and tables. |
| Formatted Text | Text without columns, retaining font and paragraph styling, graphics and tables where the target supports them. |
| Flowing Page | Recreates columns using flowing text rather than separate text boxes. |
| True Page | Retains page placement through text, picture and table boxes or frames. |
| Spreadsheet | Places each recognized page in a separate worksheet. |
Excel export with Formatted Text handles detected tables differently: each table receives its own worksheet. Other page content goes on a final index worksheet. Available layout choices depend on the selected output format.
HTML output stores graphics as separate JPEG files alongside the document. Saving an OmniPage Document, or OPD, preserves work for reopening in OmniPage. Pages can be exported repeatedly without replacing that resumable workfile.
Creating searchable PDFs
The eDiscovery Assistant processes image-only areas in a PDF while leaving existing text, metadata and annotations intact. Users can request zoning or proofreading and choose whether the searchable result overwrites the original or preserves it.
PDF Searchable Image output retains the original page appearance and adds linked text for searching. PDF Image output contains only the page images. Normal and Edited PDF output instead use the recognized text and its selected formatting.
Ultimate’s PDF options include passwords and permissions controlling copying, printing or modification. It can also create tagged or signed PDFs and append pages to an existing PDF. PDF/A output excludes encryption and embedded executable or multimedia content.
Creating forms and collecting their data
OmniPage Ultimate turns an unfilled scanned form or static form image into an electronically fillable document. Users select Form layout or draw a form zone; ordinary automatic zoning does not identify form elements.
Detected elements appear in the Text Editor, where they can be added, moved, changed or deleted. True Page export produces fillable PDF, RTF or XSN forms; HTML remains static. The receiving application fills the exported form.
Form Data Extraction handles sets of completed forms using an active PDF template that identifies the field layout. It produces CSV data with one column per form element and one row per completed form.
The completed forms must contain computer-filled or machine-filled entries; handwritten entries are excluded. Each non-image PDF form occupies a separate file. Image files and scanned forms must contain the number of pages specified by the template.
Automating incoming documents
The Workflow Assistant stores processing steps and their settings for reuse. DocuDirect adds a job name, input selection and timing instructions to a workflow. Every job includes a saving step that writes its results to the chosen destination.
- Watched folders process arriving image files. Users select folders, file types, optional subfolders and the periods during which watching runs.
- Mailbox jobs extract image attachments from selected Outlook or IBM Lotus Notes folders. They use the functioning mail application on the computer.
- Barcode cover pages identify the workflow for an accompanying scanned document or image-file group. A dedicated barcode folder handles file-based input.
Jobs can run immediately, later or at recurring intervals. Barcode cover-page jobs cannot recur. Workflows can be shared through a network location with colleagues who have OmniPage, and additional folders can hold workflow files referenced by barcode covers.
Reusing conversions in Launchpad
Launchpad provides a separate interface for repeated conversion tasks. Three tile choices specify the incoming page layout, output format and saving destination. Up to five Go-flow slots retain these combinations, with optional zoning and proofreading during processing.
Go-flows remain inside Launchpad and cannot transfer to the main workflow editor. The main application supplies wider controls for selected pages, image preparation, recognition settings and multiple outputs. DocuDirect supplies the timing controls for automated jobs.
