Improved offline signature verification method using parallel block analysis

Research in efficient automated solutions for signature verification has increased in recent years. In this paper we present a parallel computation method for off-line signature verification. A signature image is divided into some blocks. Individual threads will work on each block. The approach is based on feature extraction of every block derived from the image. Every block is represented by a set of features like number of pixels in each block, block-center and distance from the image center. Each type feature has a different impact for assigning some weightage to a block. Experimental result shows that the method eliminates random and simple forgeries and also reduces skilled forgery to a great extent.