AMLPR dataset containing 1501 photos where photo contains multiple license plates of different types of cars and motorbikes.
AMLPR Dataset
Dataset Description
This dataset was proposed in the paper “Graph Attention Network for Detecting License Plates in Crowded Street Scenes”. It includes images of different types of cars and motorbikes. Unlike existing license plate datasets, ours contain dense vehicles including different cars and motorbikes in a single image. This increases complexity of license plate detection since there are high chances of occlusion. In addition, there is degradation due to defocus of the camera on the many vechicles.
| Total Images | Total No. of License Plates (LP) | Top row: number of license plate. Bottom row: number of images | |||||
|---|---|---|---|---|---|---|---|
| 1501 | 4986 | 1 LP | 2 LP | 3 LP | 4 LP | 5 LP | >5LP |
| 113 | 225 | 562 | 340 | 192 | 66 | ||
Dataset Samples
Annotation Format
The annotation are in .txt files like:
161,261,212,261,212,284,161,284,LJ5991
12,216,59,216,59,235,12,235,J5T139
359,10,391,10,391,24,359,24,UNKNOWN
1,28,135,28,135,288,1,288,VEHICLE
138,75,268,75,268,359,138,359,VEHICLE
341,1,411,1,411,69,341,69,VEHICLE
455,43,540,43,540,288,455,288,VEHICLE
385,146,539,146,539,360,385,360,VEHICLE
The file structure is:
.
├── testset
│ ├── gt
│ │ ├── gt_img_0.txt
│ │ ├── gt_img_1.txt
│ │ ├── ...
│ ├── img
│ │ ├── img_0.jpg
│ │ ├── img_1.jpg
│ │ ├── ...
|
├── trainset
│ ├── gt
│ │ ├── gt_img_0.txt
│ │ ├── gt_img_1.txt
│ │ ├── ...
│ ├── img
│ │ ├── img_0.jpg
│ │ ├── img_1.jpg
│ │ ├── ...
License / Terms of Use
The dataset (i.e., photos) are released following YouTube’s Terms of Service. The annotations are released under Creative Commons Attribution 4.0 License
How to Cite this dataset
Chowdhury et al. "Graph Attention Network for Detecting License Plates in Crowded Street Scenes", Pattern Recognition Letters, 2020.
or
@article{chowdhury2020,
title={Graph attention network for detecting license plates in crowded street scenes},
author={Chowdhury, Pinaki Nath and Shivakumara, Palaiahnakote and Kanchan, Swati and Raghavendra, Ramachandra and Pal, Umapada and Lu, Tong and Lopresti, Daniel},
journal={Pattern Recognition Letters (PRL)},
year={2020}
}
Download this dataset
Please note, downloading this dataset means you agree to the above mentioned Terms and Conditions. This dataset is only provided for non-comercial purposes. For comercial use, please contact authors at mail@pinakinathc.me