Offline signature verification based on discrete cosine transform

Abstract

Signature identification and verification is considered among the most popular biometric methods in the area of personal authentication. The proposed method is based on offline verification of signature by two different algorithms. Before extracting different features from the signature, some preprocessing of the signature is done. In preprocessing, the signature is colour normalized and scaled into a standard format. The first algorithm is based on discrete cosine transform and it takes into consideration all the Discrete Cosine Transform (DCT) coefficients while finding match between test signature and signature stored in the database. The second algorithm is improved version of the first algorithm; it considers only significant DCT coefficients while matching. Both the algorithms use Euclidean distance classifier for comparing test signature with database. The algorithms have shown promising results while dealing with random forgeries and simple forgeries; also it gives good recognition rates.