doc_shadow_removal

This work aims to remove the shadows exiting in documents, and generate a clear and easy-to-read document.

If you try to use this code, please cite our paper "An Effective Background Estimation Method For Shadows Removal Of Document Images" which has been accepted by ICIP2019.

The test dataset comes from http://cvc.ucsb.edu/graphics/Papers/ACCV2016_DocShadow/.

Thanks the work from Derek Bradley, and the implementation https://github.com/phryniszak/AdaptiveIntegralThresholding.

We really thank Steve Bako (University of Berkeley, Santa Barbara), V. Shah (IIIT Hyderabad), Xiaoming Yu (Peking University) for interesting discussions and their code sharing.

References:

1、Bako, S., Darabi, S., Shechtman, E., Wang, J., Sunkavalli, K., & Sen, P. (2016, November). Removing shadows from images of documents. In Asian Conference on Computer Vision (pp. 173-183). Springer, Cham.

2、Shah, V., & Gandhi, V. (2018, April). An iterative approach for shadow removal in document images. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 1892-1896). IEEE.

3、Yu, X., Li, G., Ying, Z., & Guo, X. (2017, August). A new shadow removal method using color-lines. In International Conference on Computer Analysis of Images and Patterns (pp. 307-319). Springer, Cham.

4、Oliveira, D. M., Lins, R. D., & e Silva, G. D. F. P. (2013, June). Shading removal of illustrated documents. In International Conference Image Analysis and Recognition (pp. 308-317). Springer, Berlin, Heidelberg.

5、Gong, H., & Cosker, D. (2014, September). Interactive Shadow Removal and Ground Truth for Variable Scene Categories. In BMVC.