Automate Your Document Processing & Data Capture
Formerly known as OCR for Forms, OCR for AnyDoc has been used daily by more than 3,000 companies and 70,000 users worldwide in 45 countries. This robust automated solution helps businesses of all sizes improve productivity by up to 400% (with no additional staff required) while dramatically slashing the cost and time incurred by manual document processing.
OCR for AnyDoc captures data from nearly any document, and works with machine print, hand print, mark sense, bar codes, and more. Once data is extracted, you can customize business rules to match your existing business processes to validate and normalize the data prior to verification. Accurate, verified data is then seamlessly delivered to your enterprise content management (ECM), enterprise resource planning (ERP), accounting, business process management (BPM), document management, or other back-end system in a fraction of the time required for manual processing, ready for use in analysis, reporting, and retrieval.
AnyDoc OCR Benefits
- Process All Documents with One Solution: the robust ability of OCR for AnyDoc to process nearly any business document—both structured and unstructured—means you can maximize your return on investment
- Minimize Manual Data Entry & Doc Prep Costs: by eliminating manual data entry and manual document preparation, labor costs can be slashed; and even with peak document volumes, working weekends and overtime can be eliminated
- Shorten Business Process Cycle Times: data is automatically extracted and validated in seconds—and customizable verification procedures using your business rules help you ensure accuracy with minimal human intervention
- Expedite Data into Your Workflow: information from documents or document images is quickly available; with automation turning paper into usable data, you can make better and faster business decisions
- Improve data accuracy: the unpredictable possibility of human error is eliminated; using image enhancement technology, data recognition engines, consistent application of your own business rules, and database look-ups, data accuracy can reach 99%
- Eliminate Manual Document Identification: save time and effort by identifying documents automatically; robust document classification is also available
- Leverage Your Investments as Your Business Evolves: the flexible, scalable design of OCR for AnyDoc means that as your document volume grows or changes, so does the solution; OCR for AnyDoc seamlessly integrates with most front-end hardware and back-end software systems
Documents can be scanned and converted to images right out of the envelope. Scan from the mailroom, a central location, or any number of remote locations— whatever best fits your business model. Once scanned, document images are digitally viewable in seconds.
Using another system to scan your images? Retrieving them online? No problem, you can import them and still get all the benefits of automated document processing.
High image quality is crucial to identifying documents and data. A lot can happen to a document before scanning. Documents are torn. Characters are smudged or printed poorly. But OCR for AnyDoc delivers the cleanest images possible. First, scanner communication is optimized to produce the best image for extraction. Next, OCR for AnyDoc leverages years of image enhancement experience in its Quality Assure process. Images are de-skewed, lightened, cropped, de-speckled, colors are removed, and images are automatically rotated. The result is the clearest, most accurate image possible. The QA process also provides you with the abilities to replace (by re-scan) a poor quality image, insert new images into the group, or append images to the end of a scanned group.
With the cleanest image in place, OCR for AnyDoc's AutoID feature automatically identifies each document. By searching for an ID point such as existing barcodes, specific words, or matching the format of a document to others, AutoID works fast to recognize your commonly used forms and virtually eliminates the need for manual pre-sorting or separator pages.
OCR for AnyDoc examines the image to separate data from"form noise." Document layout elements such as character boxes, lines and background noise, and text are recognized as non-data and ignored. Data obscured by poorly printed characters, strange lines, and other elements go through an automated cleanup process to enhance recognition. With troublesome lines, smudges, and other unwanted marks removed and characters regenerated/enhanced as needed, OCR for AnyDoc then quickly captures data from the enhanced image.
Captured data is then validated against your business rules (best practices in computer logic form). Hundreds of built-in rules can be implemented with point-and-click ease, or you can define your own rules. Automated rules are quickly performed on data fields without operator intervention and offer an almost endless array of data integrity checks and balances. Any document that does not comply is immediately queued for quick attention in accordance to your escalation procedures and monitored until resolved. The process eliminates problem document backlogs and the costs associated with a large pile of unresolved documents.
All previous phases are designed to minimize the costs, inaccuracies, and delays associated with human intervention. At the verify stage, many documents are already 100% complete and can optionally bypass human verification, going straight to output. For other documents, operators can quickly move through extracted data fields to verify each extracted field — or can simply advance from one questionable character or business rule violation to another. You design the layout of the screen to match your business needs. Verification can take place locally, at remote locations, or both. Home workers, branch offices, new corporate acquisitions, and overseas offices can all be accommodated via the Internet.
OCR for AnyDoc also provides the ability to review the work of the verification operators. With the Auditor module, you can spot check specific items, do random checks, or check everything that an operator has seen. You will be able to compare the data before and after it has been to an operator. And this can be done either in-line (allowing for correction of operator error) prior to the data becoming output or as a post output function (which would be a review only process without correction capability).
Now that your valuable data is captured, validated, and verified, stacks of paper documents have been transformed into usable images and information. Data and images are delivered to back-end systems and put to work. With accurate information, delivered faster and at a lower cost, your organization can experience improved decision-making, shorter cash flow cycles, and better customer service, giving you a substantial competitive advantage.
The technology Driving Accurate Data Capture
The data that organizations need appears in varying formats from one document to the next. OCR for AnyDoc was developed with this in mind and will carry your data from paper to enterprise, regardless of how that data is presented on a document.
OCR for AnyDoc uses a variety of technological methods to recognize varying documents:
- Using Any Doc's AnyApp Technology, data is captured regardless of where it is located on the document, including line-item details, without the use of templates
- Using a powerful combination of methodologies, AnyApp quickly and easily locates the desired data fields and remembers their location the next time that particular document type is encountered, speeding processing
- For structured forms containing data in consistent locations, the template method of data capture provides high accuracy and exceptional speed; in a template, users define the type of data to be captured, the location of this data, and how the data will be presented on the form; OCR for AnyDoc then uses user-defined business rules to locate, analyze, and extract data with remarkable accuracy
- When occasional or seldom-seen documents are encountered, QuickApp Technology creates on-the-fly profiles almost instantly; with a simple point-and-click technology to capture data, operator interaction is reduced; a memory profile is generated to automate data capture the next time around—a big time-saver in processing speed
- With Key-from-Image (KFI) Technology's "rope and expand" capabilities, operators key significantly less data than with a manual processing system. Operators view the digital image of a scanned form and input the data on the form image into predefined fields; customizable layouts and predefined business rules ensure efficiency and accuracy
OCR for AnyDoc uses a variety of technological processes to recognize varying data types:
- Optical Character Recognition (OCR): analyzes computer-generated text located on paper documents and translates this into usable data; algorithms identify characters based upon pixilation, curvature, and line direction
- Intelligent Character Recognition (ICR): using techniques similar to OCR, ICR software analyzes and translates handwritten characters
- Voting: a technique where two data recognition engines are involved, compared, and the best solution is provided
- Optical Mark Recognition (OMR) captures responses to questions based upon a selected checkbox or bubble—a data type known as mark sense
- Barcode Recognition (BCR): reads 1D and 2D barcodes and translates them into usable data