Natural scene character recognition is challenging due to the cluttered background, which is hard to separate from text. In this paper, we propose a novel method for robust scene character recognition. Specifically, we first use robust principal component analysis (PCA) to denoise character image by recovering the missing low-rank component and filtering out the sparse noise term, and then use a simple Histogram of oriented Gradient (HOG) to perform image feature extraction, and finally, use a sparse representation based classifier for recognition. In experiments on four public datasets, namely the Char74K dataset, ICADAR 2003 robust reading dataset, Street View Text (SVT) dataset and IIIT5K-word dataset, our method was demonstrated to be competitive with the state-of-the-art methods.
@article{arxiv.1606.04616,
title = {Natural Scene Character Recognition Using Robust PCA and Sparse Representation},
author = {Zheng Zhang and Yong Xu and Cheng-Lin Liu},
journal= {arXiv preprint arXiv:1606.04616},
year = {2016}
}
Comments
The 12th IAPR International Workshop on Document Analysis Systems (DAS); The natural scene character image features used in this paper have been released at http://www.yongxu.org/Natural%20Scene%20Character%20Recognition%20Datasets.html