Multi-scale document image rectification utilising text-features

A multi-scale document image rectification method based on text-features is presented. In contrast to most used text-lines, the text-features are more robust to depict the distortions of document images with multi-column layouts, multi-type fonts or non-textural objects. Then, the rectification problem is formulated utilising a reverse strategy according to the text-features. The experiments have demonstrated the flexibility and high performance of the approach.