High-Recall 2-Dimensional Barcode Recognition Using Deep Neural Network

Activity: SupervisionCompleted SURF Project

Description

Matrix barcodes, such as QR codes and Data Matrix codes, are extensively used in industrial manufacturing. Applications of such code include product tracking, logistics management, and quality tracking. However, in industrial settings, steamy production environments, stained, tilted, and worn barcodes, blurry and overexposed images can substantially hinder conventional code scanning devices from accurately operating. This project proposes barcode recognition workflows that targets two common cases where conventional barcode scanners fail to reliably scan the barcode. Our solution has achieved considerably higher recall when compared to its predecessors.
Period1 Jun 202131 Aug 2021