Abstract—This paper introduces a new and robust model for signature recognition by means of features inspired by the human’s visual ventral stream. A feature set is extracted by means of a feed-forward model which contains illumination and view invariant C2 features from all images in the dataset. Also we use from Linear Discrimniant Analysis (LDA) to reduce the dimension of C2 feature vectors that is derived from a cortex-like mechanism. Then we utilized standard K-Nearest Neighbor (KNN) as classifier. The effectiveness of the approach is evaluated on an experimental signature database. By this new effort the rate of signature recognition is significantly high toward other models.

Index Terms—Signature Recognition, Visual Ventral Stream, HMAX, C2 features.

Reza. Ebrahim pour. Assistant Professor, Department of Electrical Engineering, Shahid Rajaee University, Tehran, P. O. Box 16785-136, Fax :+982122970006 Iran; research fields: human and machine vision, neural networks and pattern recognition.
Ali. Amiri. Master student, research fields: image processing, pattern recognition and neural networks. (E-mail: evinar@gmail.com).
Masoom. Nazari. Master student, research fields: image processing,pattern recognition and neural networks. (E-mail: innocent@gmail.com).
Alireza Hajiany, BS student, research fields: signal processing, ensemble neural networks, face recognition.

Cite: Reza Ebrahimpour, Ali Amiri, Masoom Nazari and Alireza Hajiany, "Robust Model for Signature Recognition Based on Biological Inspired Features," International Journal of Computer and Electrical Engineering vol. 2, no. 4, pp. 622-626, 2010.