Algorithm of Document Skew Detection Based on Character Vertices

We propose a new skew detection algorithm, which is based on the lowest pixels of a set of characters. The skew angle is evaluated by applying moment calculation, and the method to get characters in one text-line is described in detail. Document images of various languages, including English and Chinese, can be validly treated by this algorithm. Both the document images with large skew angle and the document images containing a few characters can be treated by this algorithm.