Redaction as a Service

Government agencies and corporations face new legislation and mandates that require the protection of individual privacy information. Removing individual privacy information from documents (redaction) can be a time consuming and costly process. The widespread adoption of document imaging systems and the growth of web-based document distribution have created a need to automate the redaction of large quantities of documents. Current economic realities have also forced government agencies and corporations to seek automated redaction solutions that limit hardware and capital investment and can be deployed rapidly without utilizing internal IT resources. The ProcessPoint "Redaction as a Service" solution was developed to address this need.

How it Works

The process starts when the server is presented with a batch of document images. The server uses world-class OCR (Optical Character Recognition) technology along with customer-specific rules to identify the words and numbers that meet your redaction criteria. A user is then presented with each image in a specialized verification interface. The system highlights all candidates and allows the user to easily approve or reject each redaction. Since no OCR system is 100% accurate, manual redaction can also be performed on any area of the image. Once verification is complete, the system will produce a new image for each redacted page with the private information completely removed and replaced with a redaction marker.

The system is configured to meet your specific redaction requirements. The system can be configured to find almost any type of data; for example the system can find and redact:

  • Social Security Numbers
  • Names of Individuals
  • Credit Card Numbers
  • Telephone Numbers
  • Any string of characters that either has a distinct pattern or can be referenced from a database
Redaction Flowchart


Our redaction technique provides a subtle long-term value over other systems. Some redaction systems will completely remove the text and leave a white space. When this is done the reader has no way of knowing if the white space once contained text or was always blank. This can alter the meaning of a legal document in some instances. Other redaction systems use a black box to obscure the text. This causes a great deal more printer toner to be used if the document needs to be printed. Our approach completely removes the pixels in the identified area, then inserts a grey redaction marker that clearly tells the reader that information has been removed.