Utilizing a Robust OCR System to Produce High Accuracy to the Character Level
Optical character recognition (OCR) is the electronic conversion and identification of scanned images, such as those from paper documents, into searchable, digital text that is readable in a PDF format. DCL’s experts can help you determine if OCR is right for your legacy materials project.
Unlike many OCR vendors who key in the data and provide minimal clean-up, DCL’s process combines a detailed manual component with robust OCR software resulting in the highest accuracy rate available and fast turn-around time. Our process includes proofreading on the character level and the ability to produce multiple outputs simultaneously. This is particularly important when the source scans are of low quality.
DCL can provide service at any level depending on the accuracy rate you require.
- Light: OCR without clean-up (accuracy level generally of 95%)
- Medium: OCR with errors cleanup (accuracy level is around 98-99%)
- Full: OCR including full proofreading - character level comparison (99.995% accurate)
- Automated: Large volume projects can be automated- system picks up and processes the file then delivers the results (99.995% accurate)
DCL is fully equipped to handle all types of scanning projects including large-scale as well as the many types of paper-based projects that require special or white-glove handling to scan, such as old and rare materials. We are experienced with both destructive and non-destructive materials.
DCL is also available to provide composition services upon request.