FormXtra Capture by Parascript, LLC
FormXtra® Capture is a complete solution for document capture, classification and recognition that delivers critical information into business processes faster and more accurately for increased productivity and reduced costs. All modules are included—capture, classification, data extraction and validation—to reduce integration requirements and related complexities. Users realize significant savings during implementation, ongoing use, maintenance and upgrades. FormXtra Capture is an enterprise-grade solution that is scalable to support medium and high-volume, complex processing needs for any size business.
“The beauty of FormXtra has always been that it processes all types of data—structured, semi-structured, unstructured, and zone-based data—and it captures everything from machine print, constrained and unconstrained handprint to complex cursive more accurately than other solutions,” said Greg Council, VP of Marketing and Product Management.
As mentioned above, FormXtra Capture processes virtually any document format and text type allowing businesses to capture and extract key data for transactions, enable workflows, automate processes and improve content organization. It provides comprehensive data capture, recognition and validation for both machine print (OCR) and handprint and cursive handwriting (ICR), barcodes and OMR. It includes Parascript’s advanced address location and recognition, check classification and recognition and signature verification technology, in one unified software package.
Parascript® Document Classification combines an easy-to-use analysis tool with a powerful production Software Development Kit (SDK) that enables organizations to efficiently and quickly identify and organize document-based information to support compliance, discovery and data management applications.
Classification is made easy with Parascript’s document classification, understanding and organizing your documents. Even if you don’t know anything about your documents, it is now simple to organize them. Once they are classified, you can add simple workflows to add metadata and even use the data within the documents to add more-descriptive searchable data. Classification employs a variety of classification methods and is based on three different types of classifiers: Content Classifier, Visual Classifier, and Auto Classifier. Content classification employs text-based features of documents. Visual classification uses “feature extraction” based upon key visual characteristics of the documents. Auto classification uses both of them and synthesizes the results for the best answer.
Intelligent Document Recognition
Intelligent Document Recognition (IDR) is a complex solution to accomplish diverse processing needs based on data use, form structure, data type, document classification, business rules workflows and application integration requirements. The Parascript IDR solution, considers business processes, content requirements, and output needs.
Parascript’s IDR solution is comprised of four major functions:
- Capture – Document content is scanned and digitized to ‘capture’ all the data in documents or forms for processing. Data is identified and extracted from various document types that can include structured, semi-structured, and/or unstructured forms or from selected data fields within these types of documents or forms.
- Classification – Classification is based upon a number of technologies and techniques including document layout, presence of images or logos, regions of data, presence of keywords or patterns, and the content itself.
- Recognition – Technologies to recognize data include OCR (machine text); ICR (constrained and unconstrained), and handprint, including cursive; 1D/2D barcodes; and OMR. Data can also be read from zones or areas and tables. Unstructured recognition employs advanced location and recognition techniques using dictionaries and pattern rules to identify data and perform extraction. Data output can be exported per the users’ requirements that can include to databases, to file systems as text or XML, as well as to content management systems such as SharePoint.
- Validation – Data validation workflows can be page centric where all the fields are presented on the page or field centric where only the fields that require review are displayed. Validation can be organized based upon specific need such as accuracy optimization, throughput optimization, or to handle security for sensitive data.