CD-LP: Compressed Domain License Plate Detection Database

- Citation Author(s):
-
Behcet Ugur Toreyin (Istanbul Technical University)
- Submitted by:
- Muhammet Sebul Beratoglu
- Last updated:
- DOI:
- 10.21227/2511-ym76
- Data Format:
- Research Article Link:
- Categories:
- Keywords:
Abstract
The dataset contains 2,400 vehicle images for license plate detection purposes. Images are taken from actively operating commercial cameras which are installed on a highway and in an entrance of a shopping mall. Images
contain generally one vehicle, but sometimes can contain two or more vehicles. For each image in pixel domain there exists two different images generated from encoded High Efficiency Video Coding (HEVC) stream using our method.
Instructions:
3 SUB SETS
•2,400 Pixel Domain Images
•2,400 HEVC Domain Images Generated from Our Block Partition Method
•2,400 HEVC Domain Images Generated from Our Prediction Based Method
TRAIN/TEST SETS
•Each train test set contains 1,800 images.
•Each test set contains 600 images.
NAMING CONVENTION
Images are given numeral names starting from 100,001 to 102,400 for each method. The same numbers are used to represent HEVC domain representations of pixel domain images.
PLATE ANNOTATION
For each image there exists another file which contains plate annotation information in YOLO format.
FOLDER STRUCTURE
+---DB
| +---HEVCDomain_BlockPartition
| | +---Test
| | \---Train
| +---HEVCDomain_PredictionUnit
| | +---Test
| | \---Train
| \---PixelDomain
| +---Test
| \---Train