Estimation of skew angle in text image analysis by sensor array processing techniques

A new signal processing method is developed for estimating the skew angle in text document images. Based on a recently introduced multi-line fitting algorithm, the proposed method reformulates the skew detection problem into a special parameter estimation framework such that a signal structure similar to the one in the field of sensor array processing is obtained. Then the well-studied techniques in that formalism (e.g., the ESPRIT algorithm) are exploited to produce a closed-form and high resolution estimate for the skew angle. A simple preprocessing stage transforms each line of text characters into a straight line of a single-pixel width. Then, a virtual planar wave propagation environment reformulates the line fitting problem into the sensor array processing framework, and high resolution signal subspace techniques are used to obtain an estimate of the skew angle. The proposed algorithm possess extensive computational speed superiority over existing single and multiple line fitting algorithms such as the Hough transform method. Details of the proposed skew detection algorithm and experimental results are presented.