Implementing an Android Application for Automatic Vietnamese Business Card Recognition
- Authors: Hung P.D.1, Linh D.Q.1
-
Affiliations:
- FPT University
- Issue: Vol 29, No 1 (2019)
- Pages: 156-166
- Section: Applied Problems
- URL: https://journals.rcsi.science/1054-6618/article/view/195550
- DOI: https://doi.org/10.1134/S1054661819010188
- ID: 195550
Cite item
Abstract
This paper presents an application for automatic Vietnamese business cards recognition. Vietnamese business cards are characterized by complex layouts, Vietnamese-English mixed texts, and diverse typesetting. The goal is to extract meaningful information from the scanned image, such as names, phone numbers, email addresses, job titles and more. The recognition algorithm proposed is divided into five parts: (1) business card processing, (2) text block finder, (3) image binarization, (4) OCR processing and (5) linguistic processing. The final Android application supports both Vietnamese and English business cards and it manages and synchronizes data on all of the user’s devices. It also allows users to edit the contact information, or delete them at will. They can call, send messages or email to contacts right in the application or share that information with others. The result has proved that our application has a definite advantage in recognizing Vietnamese business cards, or mixed English and Vietnamese when compared to existing commercial application on the Google Play store.
About the authors
Phan Duy Hung
FPT University
Author for correspondence.
Email: hungpd2@fe.edu.vn
Viet Nam, Hoa Lac High Tech Park, Hanoi
Dao Quang Linh
FPT University
Author for correspondence.
Email: linhdqse03570@gmail.com
Viet Nam, Hoa Lac High Tech Park, Hanoi
Supplementary files
