QAI was contracted by CSC to support the document imaging requirements of their customer the Nuclear Regulatory Commission. The documents converted were part of the NRC ADAMS system dating back to 1957.
The documents consisted of 35,000 separate files totaling more than 630,000 pages. The paper size used in the documents was typically 8.5 x 11 and 11×17. Most of the documents were printed simplex with an occasional duplex printed page. Clean up of the images was required in some cases. Clean up consisted of cropping the images to size, and applying image filters. Some of the documents contained halftones that required grayscale or color scanning. QAI utilized Virtual ReScan (VRS) technology to aid in the quality and output of the images. All of the documents required indexing by their unique docket number, which was then used to replace the file name for each of the documents.
Due to the age of many of the documents, variability existed with the condition of the documents, which ultimately affected the image quality and OCR results. CSC requested that the scanned documents be maintained as Group 4 CCIT TIFF files, while utilizing Optical Character Recognition (OCR) support for the flat ASCII file formats. Several voting technologies were used to increase the accuracy of the OCR output. The documents were transported from the NRC location to the QAI imaging facility located in Columbia, MD for processing. QAI implemented several methods for tracking and quality control to ensure the effectiveness and success of the conversion project.