Feature Guide

Transcript

Feature Guide FormXtra and FormXtra Capture 5.3 June 2013 Abstract Parascript FormXtra® and FormXtra Capture 5.3 provide businesses with the ability to process documents more efficiently, from a single solution capable of recognizing machine print, handprint and cursive writing. Parascript’s solution allows for structured and semistructured forms recognition support, new document identification and classification capabilities, and a highly customizable API. With FormXtra and FormXtra Capture’s new and extensive .NET interface, you can enable your solutions to incorporate many different types of business and office documents and enable simplified and scalable extraction of key data. Take advantage of Parascript’s comprehensive OCR to enable the widest variety of data support. This feature guide describes the benefits of FormXtra and FormXtra Capture 5.3. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.772.7478 F: 303.381.3101 | [email protected] www.parascript.com FormXtra and FormXtra Capture Feature Guide What’s in this Feature Guide? Contents Overview of FormXtra and FormXtra Capture 5.3 3 New in FormXtra Capture 5.3 3 Support for Structured and Semi-Structured Documents 4 Optional Modules 4 Features and Benefits - Common Features 5 Features and Benefits - Exclusive Features 7 Key Benefits FormXtra and FormXtra Capture 5.3 8 Architecture and Scalability 9 Scenarios and Examples 9 Document Processing Modules 10 Development with FormXtra 11 System Requirements 11 Summary 12 About Parascript 12 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 2 www.parascript.com FormXtra and FormXtra Capture Feature Guide Overview of FormXtra and FormXtra Capture 5.3 FormXtra is the document capture and development platform used by integrators and original equipment manufacturers that simplifies document processing with an easy-to-configure engine capable of recognizing machine print, handprint, and cursive on structured and semi-structured documents. FormXtra Capture is the document capture and data extraction solution built specifically around ease-of-use, performance scalability, and complete data-type support to capture-enable organizations of any size. New in FormXtra Capture 5.3 With the release of FormXtra 5.3 (both FormXtra and FormXtra Capture) Parascript is building upon its full-featured product with the following new capabilities and enhancements: • Support for check and “check-like” documents (e.g. deposit slips, money orders, withdrawal slips) as a built-in document recognition and data extraction module • Support for PDF input and PDF output capabilities • Built-in PDF output to SharePoint document repositories • Support for handwritten fields in semi-structured documents and forms With FormXtra Capture 5.3, any organization can streamline the processing of documents with any data type and get critical business information into business processes quicker and with greater accuracy. Additionally, dual support for both document-centric and field-centric data validation ensures that organizations can take advantage of sophisticated load-balancing, performance monitoring, automation, and field-level security to support a wide variety of data capture need. The expansive FormXtra Capture solution offers structured, semi-structured and fieldbased data processing of machine print, constrained and unconstrained handprint and cursive. This enterprise-grade solution supports high-volume, complex document processing needs, and delivers flexibility for processing rules and workflows but is easy enough to use that even smaller organizations can implement and use it without the significant investment often found with other capture products. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 3 www.parascript.com FormXtra and FormXtra Capture Feature Guide In addition to the out-of-the-box capabilities, FormXtra’s .NET API provides unprecedented control over forms processing definitions, routines and output. For users, this means a single solution that can process many types of business and office documents, and makes extracting key data simple and scalable. Support for Structured and Semi-Structured Documents Organizations may choose from two editions of FormXtra or FormXtra Capture; 1) structured or 2) semi-structured. Each is tailored to meet specific solution needs identified by customers. Please see the FormXtra Feature Comparison chart for more information on the differences between the product editions. Even so, organizations can easily move from structured forms to semi-structured forms without investment in new technologies or even introduction of new software. Access to both is managed through the same form design and administration software. FormXtra structured forms support consists of a solution for forms and documents that are uniform in data layout. The solution allows creation of templates and business rules that govern how documents are identified, data is identified, and data is extracted from documents. FormXtra semi-structured support utilizes advanced techniques designed to accommodate identification of documents and data within those documents that are less uniform or, in some cases, completely undefined. It allows identification of data elements that provide “hints” as to the type of document which then drive business rules that help located key data for performing recognition and extraction. Optional Modules While FormXtra allows developers the freedom to design and use any type of structured or semi-structured form processing, Parascript has designed pre-built form processing modules based upon its years of experience designing and supporting high-volume forms processing implementations. Organizations can take advantage of these optional modules to accelerate their own solution needs and can use them with no modifications or completely customize them. Current modules for FormXtra include health claims processing (CMS 1450 and 1500) and invoice processing. New for 5.3 is a check processing module. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 4 www.parascript.com FormXtra and FormXtra Capture Feature Guide Features and Benefits FormXtra and FormXtra Capture — common features: Feature Description Benefit Mixed Structured and Semi-Structured Document Capture NEW! Within a document type or across document Improve workflow flexibility and reduce the types, rules can include processing number of batch types that are defined and instructions for any document format. managed. Mixed Structured and Semi-Structured Data Capture NEW! On the same page, process fields, whose exact position is known, and fields that should be dynamically located. If data is always in the same location, developers can use standardized structured templates and then use location rules to provide for semi-structured data location and extraction. PDF Support NEW! Supports input and output of PDF files. Output supports direct integration with SharePoint. Provides broader support of document types common to an organization. Form Definition Versioning Automatically creates a new version of a form definition when an existing definition is changed. Simple version management allowing for users to understand and manage changes of form definitions. Semi-Structured Document Registration Use keyword and pattern-matching to identify data for recognition and extraction for any page within a batch. No need to establish specific locations or areas of a document to perform registration. Semi-Structured Data Recognition Use keyword and pattern-matching to identify data for recognition and extraction for any page within a batch. Extends the “semi-structured page” concept to all documents within a batch; No need to define specific templates for each variation of document format. Field relationships for data location NEW! Use field relationships to aid with location of data for extraction. Users can create multiple relationships to aid with field location. Increases simplicity of defining a document by allowing field location to be relative to each other. Recognition for dynamic handwritten fields NEW! Located and recognized handwriting on semi-structured forms. Provides a more-complete data extraction solution by supporting all types of form data. Dynamic OMR Support Users can create rules for dynamically locating and processing OMRs. Eliminates the need to use templates for variable forms that include optical marks. Relative Distance from Keywords NEW! Users can defined the minimum and maximum expected distance of a field from its keyword locator. Gives users more control and precision for data that is harder to capture. Dynamic Table Data Recognition NEW! FormXtra can automatically process tables of variable length and width. Less time for set-up as tables do not have to contain a fixed number of rows or columns defined. Field Type Recognition IMPROVED! Now all field types currently supported in structured forms will be supported in semistructured processing. Provides much wider support of field types for extraction. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 5 www.parascript.com FormXtra and FormXtra Capture Feature Guide Feature Description Benefit Locate Complex Field Types e.g. Addresses IMPROVED! Address fields are improved to include semi-structured location at one or more addresses on a page. Page-level OCR IMPROVED! Allows processing of all text on a page prior Versions prior to 5.x only used page level to performing field location. via 3rd party. 5.x introduces native support in addition to 3rd party options. .NET API NEW! The API is exposed as a .NET interface and expands the 50 functions to over 1500. Vendors and integrators now have very findgrained access to the powerful FormXtra core engines, including form recognition, field recognition and image processing. .NET Scripting Engine NEW! Form definitions can now use C# or VB.net. Scripting within form definitions can leverage new programming capabilities which dramatically increase productivity and simplicity. API Access within Business Rules NEW! The business rules have an access to all public classes, enumerations, methods, events, and properties of the FormXtra API. Scripts can now leverage the full power of the FX API and simplify or reduce special script coding. Full Context Access to Scripts IMPROVED! Access to all recognition context via scripting. Adds significant flexibility into workflow customization. IntelliSense for Scripting IMPROVED! Expanded support for syntax of the FormXtra API. When creating business rules or extending processing functions, you now have access to the library of FX functions directly within the scripting editor which speeds-up your work. Image Viewer IMPROVED! The image viewer within Form Definition Studio has been completely re-written and now includes new editing tools directly from the toolbar. Users now have more capabilities to refine form template images to improve document recognition. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 6 Improves the range of data types that can be processed on a form. www.parascript.com FormXtra and FormXtra Capture Feature Guide Features and Benefits FormXtra Capture — exclusive features: Feature Description Benefit Field-based Validation Administrators can set-up validation to focus users on specific fields instead of requiring the user to scan the entire document. Enables significant improvement in validation speed and accuracy. Performance-based Validation Use accuracy and speed statistics to automate routing of field validation based upon optimal performance. Enhances throughput and accuracy by allowing administrators to “tune” the data validation workflow. Real-time Statistics View in-process document capture and data validation information. Allows administrators to dynamically adjust batches and keying activities based upon specific processing needs. Real-time Validation Statistics Administrators can view statistics associated with registration, recognition, and validation of batch data. Allows administrators to understand the current effectiveness of data capture and validation. Single and Multi-pass Validation Set-up workflows to support simple single validation or more sophisticated validation schemes using two validation passes or add data auditing. For business critical data, the ability to easily create workflows to ensure data accuracy is essential. FormXtra Capture supports the widest set of validation workflows. Centralized and Automated Quality Control Administrators can set performance thresholds for batch classes to automate certain data validation workflows. Allows administrators to set batch class-specific quality thresholds and actively monitor and route them to ensure data extracted is the highest quality without increasing exception handling. Batch Prioritization Administrators can prioritize batch processing on the fly. If data associated with a certain document type needs to be processed ASAP, administrators can ensure that documents associated with the appropriate batch class is prioritized above other capture processes. User-based Field Validation Direct field validation and auditing activities based upon security or user performance criteria. Field validation concentrates validation on only specified field types. Significantly improve performance, accuracy, and security by only routing specified fields to targeted users. SharePoint Release Allows export of data into SharePoint. Auto-create fields based upon form definitions or map them to pre-existing fields in SharePoint. An easy-to-use integration that is pre-built and ready to use for businesses wishing to use SharePoint to perform workflows or as a document archive. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 7 www.parascript.com FormXtra and FormXtra Capture Feature Guide Key Benefits FormXtra and FormXtra Capture 5.3 Expansive Recognition With FormXtra and FormXtra Capture 5.3, a single solution supplies structured, semistructured, and field-based data processing of machine print, constrained and unconstrained handprint, and cursive. It also adds intelligent document recognition using the same underlying capabilities. No other document processing offering provides more in one single solution. Advanced Data Validation Workflows FormXtra Capture supports both document-centric and field-centric validation enabling routing of specific fields to identified users based upon security, validation accuracy, keying speed, and more. This enables any organization to understand key performance indicators and make real-time adjustments to maximize accuracy and efficiency. Real-time Workflow Statistics FormXtra Capture provides key managers and administrators with real-time batch and user statistics including fields successfully recognized vs. those that need to be keyed, user performance, recognition accuracy statistics and more. Business Rules Flexibility With .NET scripting and full access to recognition context, business analysts can create very rich document processing rules and workflows leveraging easy-to-use VB.NET or C# syntaxes. Additionally, scripts have complete access to the .NET API giving developers even more power to incorporate both FormXtra functions and their own in a seamless process. Enterprise-Grade Solution FormXtra provides the core document processing in-use at some of the largest capture implementations in the world. Implementations can take full advantage of multiple processing and 64bit computing to support high-volume, complex document processing needs. Integration Flexibility With both versions of FormXtra, integrators can fully-embed forms processing power using .NET languages inclusive of the form definition and execution. This means that they have unprecedented control over forms processing definitions, routines, and output. Additionally, the new help system within FormXtra 5.0 can be compiled within Visual Studio.NET to provide direct access to API guides and associated information. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 8 www.parascript.com FormXtra and FormXtra Capture Feature Guide Architecture and Scalability FormXtra is deployed as a set of Windows services that can be run in parallel to provide high-volume document processing capabilities. All processing occurs in-memory minimizing the need to utilize disk I/O which can impact overall performance. Additionally, FormXtra can maintain several hundred different document class variations within a single process which provides support for input streams containing a wide range of document types. Scenarios and Examples Multi-Document Stream Processing within Financial Services For financial institutions that have heavy document processing requirements, for instance with loan applications that require application processing, check image processing, and potentially signature authentication, FormXtra can power any document-centric or straight-through-processing solution that requires access to key data to initiate, process, or complete common financial workflows and transactions. Health Record Processing for EHR Conversions Conversion of patient diagnostic and procedural information within records is a necessity with adoption of new electronic health records and electronic medical records initiatives. Using FormXtra’s ability to locate handwritten codes used for superbills and claims, clinical software solutions can provide new capabilities to ease this transition from paperbased to electronic clinical documentation. Accounts Payables FormXtra comes with an optional module to automatically process variably-formatted data contained within purchase orders and invoices to streamline the accounts payable processes within midsize and large enterprises. Software solutions for businesses can easily embed and provide a complete paper-to-digital accounts payable solution for their customers. Automated Redaction for Freedom of Information Requests or to Protect Other Sensitive Information Using FormXtra’s ability to identify and process information located randomly on documents, solutions used by government and other industries that need to care for sensitive information contained within documents can add the ability to proactively or reactively submit documentation for automated redaction or information protection needs. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 9 www.parascript.com FormXtra and FormXtra Capture Feature Guide Document Processing Modules for FormXtra In order to get businesses up-and-running quickly or to expand capabilities, FormXtra comes with several optional modules. Document Modules Health Claims Pack The optional Health Claims Pack module provides pre-trained form definitions for both the CMS-1500 and UB-04 claims forms used by the Centers for Medicare and Medicaid. The module supplies all necessary template images, field vocabularies and rules enabling businesses to simply load the form definition and begin processing claims forms for both black-white and red drop-out forms. Invoice Pack The Invoice Pack module provides a pre-trained form definition for capturing the most common data on the invoice (typically referred to as the “header” and “footer” data). This data includes invoice number, date, purchase order number, subtotal, freight, and total. There are also default line item fields for quantity shipped, description, unit price, and net price. In most cases, businesses can load the form definition and begin to process invoices without modification. Recognition Modules Barcode Pack The Barcode Pack allows businesses to read a variety of 1D and 2D barcodes for use as data fields or document/page separators. Separate licensing required. ABBYY Voting Pack The ABBYY Voting Pack provides a runtime version of ABBYY FineReader that enables companies to compare results of FormXtra’s built-in recognition with an industryrecognized OCR package. Additionally, FormXtra will automatically apply voting to identify the best answer and merge the results on a document level insuring that only the most accurate answer for each field is output. Separate licensing required. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 10 www.parascript.com FormXtra and FormXtra Capture Feature Guide Development with FormXtra Platform for Extensibility System Requirements When designing new product releases, Parascript balances new features and functionality with providing customers the highest possible level of interoperability and/or coexistence with previous product versions. FormXtra provides the richest most detailed .NET interface to enable a fully-embedded document processing capability within any application. The following are the key functional areas supported within the API: • Image Processing: Includes functions to rotate, flip, deskew, despeckle, binarize, invert, and other common image processing capabilities. • Form Registration: Allows registration (recognizing) a document through input of the image, definition of registration zones, specifies how documents are separated, the order of pages, and captures results. • Form Definition: Allows defining a form definition inclusive of both structured and semi-structured data rules. • Recognition: Allows submission of image snippets (or fields) and associated context (such as field type, vocabularies, etc.) in addition to page-level recognition and gathers recognition results. • Data Export: Supports ASCII and XML output in both raw and structured formats, as well as direct-to-SharePoint integration. To use FormXtra Capture, you need: • Windows Server 2008 • 1GB RAM • Microsoft SQL Server 2008 Network Requirements • For some features to work properly, FormXtra servers must be a member of a Microsoft Windows 2008 domain. A gigabit per second (Gbps) network interface card (NIC) is recommended. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 11 www.parascript.com FormXtra and FormXtra Capture Feature Guide Summary Parascript FormXtra combines the greatest level of integration capabilities with the broadest support for document-oriented data including machine print, constrained or unconstrained hand print, and cursive writing. Using FormXtra, system integrators and software vendors can focus on their own solutions and core capabilities and still easily incorporate necessary business information “trapped” within common business transaction and knowledge worker documents. With the richest document processing API and support for more advanced semi and semi-structured document processing, and proven recognition technology, FormXtra is the clear choice for today’s document-oriented software solutions and those solutions that need access to document-oriented data. About Parascript Parascript is a leading developer of cursive, handprint, and machine print recognition solutions. Leveraging digital image analysis and advanced pattern recognition, its software enables business automation in forms processing, postal and financial automation, and fraud prevention; and supports cancer screening in medical imaging. Parascript’s awardwinning technology draws on a proven 15+ year track record and processes billions of document images annually. Fortune 500 companies, postal operators (including the U.S. Postal Service), major government and financial institutions rely on Parascript products, which are distributed through its OEM and Value Added Reseller networks, including partners such as: IBM, EMC, Bell and Howell, Fiserv, Selex Elsag, Lockheed Martin, NCR, Siemens, and Burroughs. Visit Parascript online at http://www.parascript.com. 6273 Monarch Park Place, Longmont, CO 80503 USA T: 303.381.3100 | toll free: 888.225.0169 F: 303.381.3101 | [email protected] 12 www.parascript.com