Automated page orientation and skew angle detection for binary document images

Abstract We describe the development and implementation of algorithms for detecting the page orientation (portrait/landscape) and the degree of skew for documents available as binary images. A new and fast approach is advanced herein whereby skew angle detection takes advantage of information found using the page orientation algorithm. Page orientation is accomplished using local analysis, while skew angle detection is implemented based on the processing of pixels of last black run-lengths of binary image objects. The experiments carried out on a variety of medical journals show the feasibility of the new approach and indicate that detection accuracy can be improved by minimizing the effects of non-textual data.