e. Document Sorter document classification Software E.DS
NTT DATA NJK

The document classification software "E.DOCUMENT SORTER" realizes the efficient document classification and sorting of a wide variety of documents (such as documents, forms, etc.) in the company as an RPA. The document classification software "E.DOCUMENT SORTER" can be automatically categorized by matching input images scanned by scanning and faxes (multifunction devices) with lines and forms (ruled lines, characters, barcodes) registered documents and forms registered. Identification can use the information on the ruled lines, characters (OCR results), and barcode/QR code, so we can respond to various formats such as receipts, invoices, applications, reports, contracts, contracts, and orders. If you use the automatic classification software "E.DOCUMENT SORTER", you will automatically achieve documentation and sort and reduce the number of operations as an RPA. A wide variety of documents can be classified By freely registering a model of the character information, ruled line information, and barcode information, and a model, a variety of documents and forms can be categorized into a high -precision folder by freely registering AND and or functions. Document classification document examples "Specialized books / academic books", "newspapers / magazines", "textbooks / questions", "specifications / manuals", "contract / business reports", "business reports", "general conference", "board of directors", "books" "Reception and work instructions", "Work daily report / process list", "attendance management slip / time card", "questionnaire / consultation card", "order form / application form", "questionnaire / application postcard" "Entrance application / course delivery" ... etc. Document classification by OCR identification OCR is equipped with a high -performance recognition engine that has a proven track record in the type OCR software (OCR recognition language is Japanese/English). High -speed recognition (match / partial match) of characters within the specified range. Document classification by ruled line identification Equipped with a high -precision ruled algorithm that captures the ruled line information by the characteristics of lines and intersections. It is also possible to specify not only documents and forms, but also some areas as comparisons. Extract the square frame, L -shaped, cross -shaped ruled lines, and output results by modeling the template. * The ruled line is recommended for black or dark concentration. Barcode identification document classification Equipped with high -performance recognition engine that supports low quality image quality. Barcode (JAN, ITF, Code39, Code128, NW-7) and QR code are supported. * Not supported by micro QR. Multicore compatible (Server OS version only) The multi -core environment (an environment equipped with two or more processor core) is equipped with a function to handle each core at high speed. * This function is supported only for Server OS version. Image correction In order to improve document classification and identification, an image correction function such as straightforward correction, inclination correction, noise removal, etc. that correct the read image in advance is incorporated. Rename function Outputs are possible with any file name for images classified after identification. Other features Multiple folder monitoring The image folder for classification can monitor up to eight folders. Multi page TIFF compatible The following processing is possible for multi -page TIFF. ･ Identify and classify each page. ･ Only one page is identified, and collectively and classify. ･ Separated before and after the model registration form, output in another file (division output) Rejected function Unbelievable documents and forms can be output to any specified folder as a reject process. File notes It is possible to identify with the file name. Log output function Detailed identification processing history such as image correction contents, OCR results, and code identification results can be output.