Barcode Image Enhancement Using Deep Learning

The use of barcodes to store alphanumeric information has spread from supermarkets to department stores, the health industry and even on the back of our drivers’ license. Recent years have also seen demand to increase the density of information encoded in barcodes. A Drivers’ License has its personal data encoded in the barcode on the back of the document, and the data embedded is often used as part of an identification/verification process. However, the camera captured picture of the barcode is susceptible to quality degradation and even slight perturbances in the image quality of the barcode (e.g. out-of-focus blur, motion blur, non-optimal lighting) causes failures in its readability. In this project, we aim to use artificial intelligence techniques (specifically deep learning) to improve the quality of Drivers’ License 2D barcode captured with the mobile device cameras in a way that improves its readability for identification or verification purposes.

Faculty Supervisor:

Yalda Mohsenzadeh

Student:

Vahid Reza Khazaie

Partner:

Applied Recognition Corp

Discipline:

Computer science

Sector:

Information and cultural industries

University:

Western University