User manual ABBYY SOFTWARE FORMREADER-AUTOMATED FORMS PROCESSING

[. . . ] . 22 Organizing Automated Forms Processing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Using ABBYY Technologies to Solve Untypical Tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Automated Forms Processing Form Types Introduction In the course of our lives we fill in hundreds of forms applica tion forms, questionnaires, insurance claims, etc. At the same time computers have become indispensable for collecting and managing information, making the task of extracting data from printed docu ments even more pressing. [. . . ] In terms of throughput, scanners can be divided into the following three groups: office low throughput scanners, office medium throughput scanners, and production scanners with very high throughput. Low end office scanners have a throughput of up to 500 pages per day while productions scanners may scan more than 20, 000 pages per day. If the scanner accidentally takes in two form pages at a time, this may result in some of the pages not processed at all. To prevent this from happening, many scanners have a spe cial control mechanism which weighs paper sheets, measures their thickness or measures the light that may penetrate them. printed on different kinds of paper, have different colours and paper thickness, etc. ). The solution is to use ultra sonic sensors which make sure that the signal has been reflected only from one surface, i. e. Some scanners may have a number of additional features which may also come in useful An endorser or a built in printer that prints on scanned documents an index that is then used to identify them; A hardware image enhancement module; A hardware image compression module; Colour lamps that can remove certain backgrounds (so called drop out colours red, blue or green); Caching images in the scanner's onboard memory, which makes for faster scanning. Automated forms processing: step by step Personnel training Working with ABBYY FormReader requires minimum special knowledge and training. The data capture system is usually run by operators responsible for entering data from forms and an administrator who sets up and monitors the system. Depending on how data capture is organized, the operator's job can be of two kinds all operations are performed on one computer, the opera tor loads forms into the scanner, and oversees the scan ning, recognition, and verification processes; in the case of ABBYY FormReader Enterprise Edition, dif ferent operators are responsible for specific processes scanning, document assembly, verification, and export. In the case of ABBYY FormReader Enterprise Edition, the administrator deploys the system, allocates processing roles to operators, cre ates templates and descriptions of multi page documents. The operator also oversees the flow of information within the sys tem It takes from several hours to 2 3 days to train the operators and the administrator*. All the practical knowledge required to operate FormReader can be acquired within this time period thanks to a carefully thought out training course * The administrator's training course includes the following topics : 1) production capture; 2) designing new forms; 3) creating form templates; 4) installing ABBYY products (including network installations); 5) setting up scanning, recognition and verification options; 6) allocating processing roles to operators; 7) creating validation rules and document assembly rules; 8) monitoring the operation of the program and creating reports. Processing cycles For a better understanding of how ABBYY FormReader works, let us take a closer look at the main processing cycles. A batch is a collection of similar documents which must be processed and saved both as images and as text data captured from the fields. Batches can be opened either by an operator or automatically by the program. Images of forms that need to be processed may be added to a batch in one of the following ways: by scanning paper forms; by adding pre scanned images from a special dialog box; by dragging and dropping document icons in Windows Explorer. Recognition is an automated process whereby the text in the data fields is "read" by the program and con verted into electronic form. First, the program selects the right template for the form and detects blocks from which data have to be captured. Once all the images in a batch have been recognized, some pages may contain characters about which the program is unsure. Similarly the operator corrects any errors detected by validation rules (the program marks pages with errors with special colour flags). Finally, verified and validated data are saved to a file or exported to a database. All the operator needs to do is to click the "Export" button. Throughout the entire data capture process the involvement of the operator is kept to a minimum. [. . . ] To achieve this goal, a formalized description of each field is created in FlexiCapture Studio fields are described through their various parameters and surrounding objects, such as captions, images, separators, blank spaces, etc. This description is tested on a large number of forms, and then adjusted and entered into ABBYY FormReader or ABBYY FineReader Engine as any template. Automated Forms Processing Capturing data from forms that are not machine readable Sometimes it is not necessary to capture all the data available on a document. The program creates a unique index on the basis of these fields and converts the document images into a suitable storage format. Next time when someone looks for a particular page, they will be able to find its image by carrying out a search in the index. [. . . ]