Neoptec - Specialists in Software for Automatic Document Capture by Scanner

An API consists of a group of routines, protocols and tools that computer programmers can use for integrating specific applications into their own software developments. Thus, an API is intended for incorporation into new software.

NEOPTEC uses exclusively its own software engines to carry out functions for image-processing and for OCR and ICR character-recognition.

Technology by which information is collected for computer processing, automatic data-capture makes it possible to read data not only from web-based forms but also from paper documents, such that this data is available for use in computer applications.

Automatic data-capture takes advantage of an array of technical advances in image processing and the use of neural networks to constitute a sort of software intelligence. This technology enables NEOPTEC software to capture, analyse, verify and export data read from the images acquired using a document scanner.

The kinds of objects that can be recognised with automatic data-capture applications vary according to the recognition motors used. The motors used in NEOPTEC software, for example, can recognise not only machine-printed characters, hand-printed characters (OCR and ICR) and barcodes, but also several other different types of objects, such as value scales, Sema-code, CMC7 codes (used for European check stubs), etc.

To learn more about our universal application for data capture from documents, check out DATA-SCAN. NEOPTEC has also developed serveral automatic data-capture applications specifically for assessment, including QCM Direct, NEMO-SCAN et Transcript.

Character recognition is the process by which symbols are identified during digitalisation by a scanner and processed for use in computer applications. There are two types of recognition: OCR for machine print and ICR for hand print.

The term "character recognition" is sometimes used more loosely to refer to the process of reading all the different types of data that can be recongnised using Automatic Data-capture technologies.

The recognition motors used in NEOPTEC software can read a wide variety of objects and any symbol used in western writing and typeset fonts.

A term designating the science of assessment, docimology focuses on factors that can have an affect on the way in which tests and other types of assessments are evaluated. Docimological studies have shown, for example, that there is often a considerable gap between the scores attributed to the same candidate's test paper by different correctors when a test calls for open-ended answers.

In light of the effects revealed by such studies, different strategies for tying to ensure a greater degree of objectivity in assessment have been adopted in various countries. Solutions like having two correctors check each exam, adopting an anonymous marking protocol and using closed-questions formats (multiple-choice questionnaires, some types of short-answer tests, etc.) have helped to reduce or eliminate some forms of bias in assessment.

In addition, technological advances have made it possible to automate the correction of certain types of tests, particularly MCQs. As a result, well-designed MCQ tests can be a particularly effective method of assessment for comparing a large number of candidates according to a fixed set of criteria. It is for this reason that the multiple-choice format (which nonetheless remains complementary to other methods of evaluating candidates' knowledge and abilities) plays such a key role in assessment processes requiring scoring criteria that are as clear and objective as possible.

One of NEOPTEC's key aims in developing a new-generation tool for automatically correcting tests, was to restore to education professionals their full importance in desiging intelligent assessments by empowering them to take full advantage of this technology without the constraints of earlier techniques for automatic correction.

ICR, a form of software intelligence, is the process by which an application recognises the alphanumeric characters written by hand on a paper doucment. An ICR application carries out recognition by processing a digital image acquired via a document scanner.

This technology plays a central role in the development of solutions enabling intelligent automation of data entry from different types of documents that are structured (and therefore able to be optimised for perfect recognition). It is usually used in conjunction with other functions as part of a complete Automatic Data-capture solution, making it possible to read information from forms, questionnaires, surveys and sorts of structured and semi-structured documents.

According how the writing in a document is structured and how the fields for recognition are defined, the success rate for recognition by an ICR application can be higher or lower. The system used in NEOPTEC's universal OCR / ICR application, has been copiously tested in comparison to manual data-entry and has systematically demonstrated superior performance, both in terms of speed and accuracy, to that of a professional data-entry technician .

This form of software intelligence forms the foundation of Automatic Data-capture technology. OCR is the process by which an application recognises the alphanumeric characters appearing on a machine-printed document and transforms these characters into editable digital symbols (ASCII, for example). An OCR application carries out recognition by processing a digital image acquired via a document scanner.

The OCR recognition motors used in NEOPTEC software operate data entry of printed characters of all fonts (latinate letters, numerals, symbols). See also ICR.

OMR is a technique using a dedicated machine, called an OMR scanner or optical-mark reader, to tally marks made on paper printed with special inks. OMR scanning made possible a first big step in large-scale automation in the processing of certain types of documents, particularly standardised forms and multiple-choice tests.

Still used in many universities and colleges, hospitals and government agencies, especially in Anglo-Saxon countries, OMR is progressively being replaced by more recent technologies, which are often more flexible, more reliable and more precise.

What is most commonly known as Automatic Data-capture, for example, is a technology making it possible to read a wide variety of relatively complex documents using an ordinary scanner (often known as a "document scanner" or "image scanner"). This technology uses modules integrating artificial intelligence to recognise not just special marks, but also machine-printed and hand-printed characters appearing on documents created with widely used word-processing software and printed on ordinary paper with any laser printer.

Unlike OMR scanning, automatic data-capture technologies carry out recognition using high-resolution images acquired by scanner. The acquired images can be archived and made available for later viewing. This means that using automatic data-capture allows for verifying the information automatically acquired on the source image without having to sift through all of the originally scanned documents.