Signature recognition

About: Signature recognition is a research topic. Over the lifetime, 2138 publications have been published within this topic receiving 37605 citations.

Multi Biometric System for Verification with Minimum Training Data.

Abstract: This paper presents the multi biometrics system for identity verification based on face and signature. The proposed system is designed for applications where the training database contains one face and one or two signature image for each individual. This system has three modules namely face recognition, signature recognition and multi-biometrics. In face recognition, first the face is detected using the triangulation algorithm and then it is recognized based on KDDA and the Haar wavelet algorithm. In signature recognition, the signature is matched with stored database image using the Haar wavelet. Multi-biometrics algorithm considers the results of face and signature recognition and gives the final matching result based on the fusion rule. This system is tested on a database prepared by the authors and the overall accuracy of the system is found to be 94.37%.

Off-Line Signature Recognition Based On Angle Features and GRNN Neural Networks

Abstract: This research presents a handwritten signature recognition based on angle feature vector using Artificial Neural Network (ANN). Each signature image will be represented by an Angle vector. The feature vector will constitute the input to the ANN. The collection of signature images will be divided into two sets. One set will be used for training the ANN in a supervised fashion. The other set which is never seen by the ANN will be used for testing. After training, the ANN will be tested for recognition of the signature. When the signature is classified correctly, it is considered correct recognition otherwise it is a failure. Keywords—Signature Recognition, Artificial Neural Network, Angle Features.

A Growing Cell Neural Network Structure for Off-Line Signature Recognition

Abstract: The signature recognition is a topic of intensive research due to its great importance, among others, in the financial system. However it does not exist yet an enough reliable method for signature recognition and verification, especially in the forgeries detection. This paper presents an off-line signature recognition using features extracted from the off-line signature and an array of five growing cell neural network. The proposed system was evaluated using 950 signatures of 19 different persons. Experimental results show that proposed system achieves a fairly good recognition rate with a relatively low computational complexity.

The Action Recognition Technology Based on Motion Vector Analysis

Abstract: A technology of the motion-recognition based on motion vector analysis was proposed by improving the algorithms of human behavior recognition of traditional methods of template matching.The recognition effect of this technology was verified by experiments.The technology uses the percentage of vector of standards body motion as a template.Recognition action will be to identify by comparing the percentage of vector and the known template.The technology can correctly recognize shaking head,shaking head and nod.If recognition action can be repeated three times,the recognition rate of this action will be more than 95 percent.During conduting real time action,the advantages of this technology is effective identification,simple algorithm,fast recognition and strong anti-jamming.

Biometric Features Selection with k-Nearest Neighbours Technique and Hotelling Adaptation Method

Abstract: The presented work focuses on the method of handwritten signature recognition. For those objects recognition process can be difficult because repeatability of signature features is low. The most important advantage of the proposed solution is individual adjustment of features and analysis methods which distinguish each single signature inside the dataset of signatures’ features. In presented approach different features and similarity measures can be freely selected. Additionally, selected features and similarity measures can be different for every person.