CNN-Based Text Image Super-Resolution Tailored for OCR

Since low-resolution images may hamper the performance of optical character recognition (OCR), text image superresolution (SR) has become an increasingly important problem in computer vision. The previous works concern more on the objective quality (e.g. PSNR) rather than the OCR performance. In this paper, we developed text image SR method to help OCR. In detial, we propose an edge-based loss function for SR training and conducted model combination to further improve the performance. Also, we propose a simple yet effective image padding method to refine the image boundaries during SR.

Paper | Slides

Citation:

@inproceedings{zhang2017cnn,
  title={CNN-based text image super-resolution tailored for OCR},
  author={Zhang, Haochen and Liu, Dong and Xiong, Zhiwei},
  booktitle={IEEE Visual Communications and Image Processing (VCIP)},
  pages={1--4},
  year={2017}
}