Cooperative Self-Organizing Maps for Consistency Checking and Signature Verification

Automatic Signature Verification: The State of the Art

“…In Table V, Abu-Rezq and Tolba [2] used a neural approach for signature verification based on moment invariant features and projection-based features. Bajaj and Chaudhury [15] used different types of global features: projection based (horizontal and vertical projection) and contour based (upper and lower envelope).…”

“…Table IV shows some of the NN models that have been used recently: Bayesian NNs [30], [351], multilayer perceptrons (MLPs) [7], [15], [17], [126], [167], [345], [350], time-delay NNs [22], [167], ARTMAP NNs [215]- [217], backpropagation neural networks (BPNs) [13], [15], [47], [66]- [68], self-organizing maps [1], [2], and radial basis functions (RBFs) [13], [109], [203], [232], [316]. Fuzzy NN, which combine the advantages of both NNs and fuzzy rule-based systems, has also been considered [102], [270], [353].…”

GloveSignature: A Virtual-Reality-Based System for Dynamic Signature Verification

“…In [1], a new approach is described for checking the consistency of biometric databases, and a special application to signature recognition is given. A neural-network-based consistency measure is proposed to quantify the intra- variability of the individual's signatures.…”

“…The learning phase of the feature map is an iterative procedure in which each iteration has three steps: the presentation of a randomly chosen input vector from the input space, the evaluation of the network, and an update of the weight vectors. Details of the SOFM training algorithm are to be found in [1]. Neighboring neurons in the network represent neighboring locations in the feature space.…”

“…In this paper we describe our approach to overcoming the difficulties of DSR using neural networks. Self-organizing maps have already proven themselves to be appropriate and efficient for signature recognition [1]. However, the extensive amount of data involved in DSR requires a different approach.…”