SKEW ANGLE DETECTION OF DIGITIZED INDIAN SCRIPT DOCUMENTS

Skew angle detection of scanned documents containing most popular Indi an scripts (Devnagari and Bangla) is considered. Most characters in th ese scripts have horizontal lines at the top, called headlines. The ch aracter head lines mostly join one another in a word and the word appe ars as a single component. In the proposed method the components are a t first labeled. The upper envelope of a component is found by columnw ise scanning from an imaginary line above the component. Portions of u pper envelope satisfying the properties of digital straight line are d etected. They are clustered as belonging to single text lines. Estimat es from individual clusters are combined to get the skew angle. Apart from accuracy and efficiency, an advantage of the method is that chara cter segmentation and zone detection can be readily done from head lin e information, which is useful in Optical Character Recognition approa ches of these scripts.