A Generalized Framework for Recognition of Expiration Date on Product Packages Using Fully Convolutional Networks

Introduction

This project proposes a generalized framework for detecting and understanding the expiration date of the products. The proposed framework can handle challenging expiration date cases and distinguish 13 different date formats. Moreover, it can detect and recognize expiration dates even when the input image contains multiple dates. It generates bounding boxes with predictions in the image.

Figure 1: Qualitative results of the expiration date understanding.

Executable files

We've released executable files of the proposed framework. Anyone who would like to test their images for recognition and understanding the expiration date can use our demo file.

Dataset

In the lack of a publicly available dataset, we create a novel dataset, ExpDate, consisting of date images with challenging cases. It is now publicly available for further research.

Details

Please check the project page for more details about executable files and the dataset.

Citation

If you find this dataset useful for your research, please cite:

@article{seker2022generalized,
  title={A generalized framework for recognition of expiration dates on product packages using fully convolutional networks},
  author={Seker, Ahmet Cagatay and Ahn, Sang Chul},
  journal={Expert Systems with Applications},
  pages={117310},
  year={2022},
  publisher={Elsevier}
}

acseker / ExpDateWebsite