Ship License Plate Recognition Benchmark: SLPR-R and SLPR-P

Ship license plate recognition (SLPR) plays important role in intelligent waterway management, but few attention has been paid to SLPR in scene text recognition (STR) community. We collected ship license plate text images and classified them as SLPR-R and SLPR-P according to the sampling strategy. Specifically，SLPR-R contains 6,922 text samples, of which 5,131 for training, 1,791 for testing. The SLPR-R dataset covers Chinese and digital sequence images, all of which are ship license of river boat. Noting that some practical factors lead to duplication of text content between partial images. The only difference from SLPR-R is that the text images in SLPR-P almost all have perspective, tilt, rotation and other distortions. The corpus distribution of SLPR texts shows obvious imbalance characteristics, the following figures illustrate the typical images in SLPR-R and SLPR-P datasets, as well as the top-20 corpus and their distribution in number. Extremely uneven distribution will seriously affect the generalization of the recognition model.

Please contact   (jwcao@hdu.edu.cn)us to get access to the dataset, before this please check and agree the term of use and license.

This Dataset is made freely available to academic and non-academic entities for non-commercial purposes such as academic research, teaching, scientific publications, or personal experimentation. Permission is granted to use the data given that you agree to our license terms bellow: 1. That you include a reference to our paper in any work that makes use of the dataset. 2. That you may not use the dataset or any derivative work for commercial purposes.