Vietnamese word processing and image recognition software
1VS Company and Dong Kinh Investment and Development Joint Stock Company have just announced the completion of the project to integrate the "1C: Document Management (ECM)" software and the ABBYY Vietnamese image recognition product.
The result is the creation of an integrated product suite that combines document management features with the ability to recognize images containing Vietnamese text and convert them into standard text format. Furthermore, it allows for barcode recognition to automatically classify documents. This product suite effectively assists in document digitization and the development of electronic archives.
Image recognition technology for converting text formats has been developed since the 1990s, including several open-source technologies such as CuneiForm. Previously, this open-source recognition feature was available in the 1C:ECM solution, but it was limited to recognizing only a few common languages worldwide such as English, Russian, French, Spanish, etc., and not Vietnamese. Currently, Vietnamese recognition technology is being developed by ABBYY (Russian Federation), and the integration of ABBYY technology with 1C:ECM allows for the resolution of the Vietnamese recognition problem in document management solutions.
When implementing electronic document management and archiving software systems, businesses often need to import a huge volume of pre-existing paper documents into the system. Each paper document typically corresponds to a document tag in the program. Document attribute information is entered into the document tags, and attached are files containing scanned images of the original documents. Creating document tags in the software and scanning paper documents can be very time-consuming; for example, processing 100,000 documents can take up to 300,000 minutes, or 5,000 working hours, equivalent to 625 man-days.
The integrated solution between 1C:ECM and Abby software features a completely new data import method, allowing for a reduction in image scanning and data import time by tens of times. For example, this task can be reduced to just a few weeks.
The process is quite simple. First, the solution automatically generates all text tags with a predefined structure from an Excel file. This takes less than half an hour if an Excel file is already available, and the 1C:ECM software automatically generates a series of corresponding barcodes (if barcodes are not already present). The user's remaining task is to paste the barcodes onto the documents and scan hundreds of thousands of these documents simultaneously. The average speed of an industrial scanner is 25 documents per minute, and it takes 4,000 minutes to scan 100,000 documents, equivalent to 8 working days. At that point, the 1C:ECM solution, thanks to its integration with ABBYY, will recognize the documents via barcodes and automatically attach them to the corresponding text tags.
More specifically, when receiving image files (BMP, TIF, TIFF, JPG, JPEG, PNG, GIF formats) from scanners, fax machines, etc., the 1C:ECM + ABBYY system will recognize these files and convert the images into plain Vietnamese text formats (TXT, DOC, DOCX, HTML). This feature works with both Unicode and VNi character sets. Simultaneously, the converted text will be indexed, allowing for full-text searching similar to Google. This function is particularly useful for businesses with large archives. The 1C:ECM search mechanism will browse a content index with links to relevant documents, instead of laboriously opening each document in hundreds of thousands of stored files to find the desired content.
According to Hanoi Moi