Overview

FPT.AI Reader - Vietnamese Passport Recognition

Introduction

Passport Recognition API enables effortless extracting information from passport images with the state-of-the-art Optical Character Recognition (OCR) and Deep Learning technology

Application

Passport Recognition API can be applied for automated data input process in various domains including finance and banking, telecommunications, travel, tourism and hospitality; in such use cases as bank account opening, credit card application processing, user registration, guest verification / reservation, off-site check-in, etc.

System

Features

The main function of the system is to identify and extract information from the Vietnamese Passport photos.

Users can send Vietnamese Passport images to the system, and this system will return extracted information from the photos, for example: Passport number, name, date of birth, address, etc.

In order to evaluate the trustworthy of the results, the probability of each information is also returned

Performance

The processing time of the system depends on the resolution of the image submitted by the user.

The average processing time is:

• Server without GPU: 4-5 seconds/ photo.
• Server with GPU (NVIDIA Geforce GTX 1080Ti): 0.5-1 sec / photo, process about 2 images per second.

Accuracy

The average accuracy of the system is up to 96% if the input image meets all requirements.

The detail accuracy for each information fields is:

• 96% for extracted information of Passport number, date of birth, place of birth, gender, date of issue and expiry date
• 93% for extracted information of name and ID number