Abstract
This paper proposes a method to uniquely identify commercial products with an item level barcode, specifically JAN/EAN/UCC, with which we usually cannot distinguish individual products. This research is motivated by the industrial and consumer needs for first-in-first-out operations of their inventories in the environment, where serial level barcodes, such as GS1-128 and SGTIN, or RFID are still not available. The proposal, referred to as Barcode Fingerprinting, utilizes the microscopic features of printed barcode stripes to discriminate each other. We zoom in a particular portion of a barcode image and apply SURF Algorithms and RANSAC Algorithms, to extract the microscopic features. The features are stored with the barcode ID and the time stamp for the future comparisons. We examined the feasibility of the proposal method by processing 100 barcodes of single brand PET bottles. The experiment reveals that we can uniquely identify each of the 100 barcodes by using Barcode Fingerprinting with a proper preprocessing of images.
| Original language | English |
|---|---|
| Title of host publication | IEEE World Forum on Internet of Things, WF-IoT 2018 - Proceedings |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 416-420 |
| Number of pages | 5 |
| Volume | 2018-January |
| ISBN (Electronic) | 9781467399449 |
| DOIs | |
| Publication status | Published - 2018 May 4 |
| Event | 4th IEEE World Forum on Internet of Things, WF-IoT 2018 - Singapore, Singapore Duration: 2018 Feb 5 → 2018 Feb 8 |
Other
| Other | 4th IEEE World Forum on Internet of Things, WF-IoT 2018 |
|---|---|
| Country/Territory | Singapore |
| City | Singapore |
| Period | 18/2/5 → 18/2/8 |
Keywords
- feature matching
- fingerprint
- image identification
- item management
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Networks and Communications
- Computer Science Applications
- Hardware and Architecture
- Information Systems and Management
- Safety, Risk, Reliability and Quality