Run OCR on documents

Use Optical Character Recognition (OCR) to translate text locked in document images into searchable text. OCR identifies characters using pattern recognition. Redacted areas in the document image block the text underneath from OCR.

Use the Search Builder to filter documents that contain images. To use the Search Builder, refer to Use Search Builder.

Perform the following procedure to OCR documents.

1. In the Project page, click REVIEW.
2. Choose to OCR selections in the grid or a document population (saved search, folder, or tag).
   • To OCR selections on the grid, perform the following tasks.
     1. Select the documents to OCR.
     2. In the Document Actions pane, click OCR.
   • To OCR a document population, perform the following tasks.
     1. Click and select OCR.
     2. In the OCR dialog box, in Document Population, select Saved Search, Folder, or Tag. Then, select the name.
3. In Job Name, type a name to identify the job.

4. In Language, select the language used in the document images. Available languages include: English, Japanese, Chinese Simplified, Chinese Traditional, Spanish, Portuguese, Arabic, Hindi, Indonesian, Russian, German, Italian, Korean, French, or Hebrew. 

5. In Text Destination, select OCR Text or Extracted Text. If you do not make a selection, this defaults to OCR Text.

6. By default, to prevent the redacted text from displaying in the OCR text, OCR for Redacted Text is selected. Uncheck this option only if you want to create OCR text that ignores existing redactions.

7. Click Submit.

8. To view details about the submitted job, In the Project page, click Jobs Overview.

9. In the Jobs Overview page, when the Job Status indicator reaches 100%, in the Job Name column, click the job you submitted. The resulting Information dialog box contains details about the job, including the number of documents that were selected and if any documents were skipped. For more details about this Information dialog box, refer to View OCR job details.