Skip to content

UFPR-ADMR-v2: a dataset for image-based automatic meter reading containing 5,000 dial meter images obtained on-site by employees of the Energy Company of Paraná (Copel), which serves more than 4 million consuming units in the Brazilian state of Paraná.

Notifications You must be signed in to change notification settings

guesalomon/ufpr-admr-v2-dataset

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 

Repository files navigation

UFPR-ADMR-v2 Dataset

The UFPR-ADMR-v2 dataset contains 5,000 dial meter images obtained on-site by employees of the Energy Company of Paraná (Copel), which serves more than 4M consuming units in the Brazilian state of Paraná. The images were acquired with many different cameras and are available in the JPG format with 320×640 or 640×320 pixels (depending on the camera orientation). More details are available in our paper.

Here are some examples from the dataset:

The dataset is split into three subsets: training (3,000 images), validation (1,000 images) and testing (1,000 images). Every image has the following annotations available in a .txt file: the counter’s corners (x1, y1), (x2, y2), (x3, y3), (x4, y4). The corners can be used to rectify the counter patch and represent, respectively, the top-left, top-right, bottom-right, and bottom-left corners. For each dial, the current position (x, y, w, h) and the corresponding reading (the final reading as well as the approximate reading with one decimal place precision). All counters of the dataset (regardless of meter type) have 4 or 5 dials; thus, 22,410 dials were manually annotated.

The full details and statistics regarding the dataset are available in our paper.

How to obtain the dataset

The UFPR-ADMR-v2 dataset is the property of the Energy Company of Paraná (Copel) and is released only to academic researchers from educational or research institutes for non-commercial purposes.

To be able to download the dataset, please read carefully this license agreement, fill it out and send it back to Professor David Menotti (menotti@inf.ufpr.br). The license agreement MUST be reviewed and signed by the individual or entity authorized to make legal commitments on behalf of the institution or corporation (e.g., Department/Administrative Head, or similar). We cannot accept licenses signed by students or faculty members.

In general, you will receive a download link within 1-5 business days. Failure to follow the instructions may result in no response.

Citation

If you use the UFPR-ADMR-v2 dataset in your research, please cite our paper:

  • G. Salomon, R. Laroca, D. Menotti, “Image-based Automatic Dial Meter Reading in Unconstrained Scenarios,” Measurement, vol. 204, p. 112025, 2022. [arXiv] [Science Direct]
@article{salomon2022image,
  title = {Image-based Automatic Dial Meter Reading in Unconstrained Scenarios},
  author = {G. {Salomon} and R. {Laroca} and D. {Menotti}},
  year = {2022},
  journal = {Measurement},
  volume = {204},
  pages = {112025},
  doi = {10.1016/j.measurement.2022.112025},
  issn = {0263-2241},
  keywords = {Automatic Meter Reading, Dial meters, Pointer-type meters, Deep learning, Public dataset},
}

You may also be interested in the conference version of this paper, where we introduced the UFPR-ADMR-v1 dataset:

  • G. Salomon, R. Laroca, D. Menotti, “Deep Learning for Image-based Automatic Dial Meter Reading: Dataset and Baselines” in International Joint Conference on Neural Networks (IJCNN), July 2020, pp. 1–8. [IEEE Xplore] [arXiv]

Additionally, consider showing your support by starring (:star:) this repository.

Related publications

A list of all papers on AMR published by us can be seen here.

Contact

Please contact Professor David Menotti (menotti@inf.ufpr.br) with questions or comments.

About

UFPR-ADMR-v2: a dataset for image-based automatic meter reading containing 5,000 dial meter images obtained on-site by employees of the Energy Company of Paraná (Copel), which serves more than 4 million consuming units in the Brazilian state of Paraná.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published