Automatic Signature Verification using Segmentation

Asraful Haque presents a new offline signature verification technique based on dividing the signature image into blocks. 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 feature type has a different impact for assigning a certain weight to a block. The system is easy to implement since its requirements are very feasible. In the last few decades, many approaches have been developed to verify off-line signatures automatically. Unfortunately, these concepts are under-utilized as they do not achieve a good classi?cation accuracy that is acceptable in real life applications. Haque‘s method has the potential to become a framework for a fast and accurate automatic signature verification system. Dynamic properties of this verification system make it more powerful and secure, because no single attribute of the verifiers is fixed. Center of the image and tolerance level in distance varies from image to image. Experimental results show that the method eliminates random and simple forgeries and reduces skilled forgery to a great extent. The new technique is useful for any organization like a bank where a large amount of signatures has to be verified manually. The accuracy of the system is 92.92% which is quite better than any existing system. According to Haque, there is still room for future modification in this technique. The book is especially useful for students or researchers interested in pattern recognition.