A new approach to signature recognition using the fuzzy method

Abstract

The paper presents a new fuzzy approach to off-line handwritten signature recognition. The solution is based on characteristic feature extraction. After finding signature's center of gravity a number of lines are drawn through it at different angles. Cross points of generated lines and signature sample, which are further grouped and sorted, are treated as the set of features. On the basis of such structures, obtained from a chosen number of learning samples, a fuzzy model is created, called the fuzzy signature. During a verification phase the level of conformity of an input sample and the fuzzy signature is calculated. The extension in feature extraction as well as proposed fuzzy model has never been employed before. It needs to be emphasized that information stored within the verification system cannot be used to recreate the original signatures collected at the enrolment phase. The fact is particularly valuable for large databases and systems where storage safety is crucial. The solution is very flexible and allows the user to extend an intuitive structure of fuzzy sets by employing dynamic features, making the approach an on-line method. The results obtained should be still improved, similarly to the case of other known biometric systems related to signature recognition. However, the presented technique can be easily utilized in applications where FAR coefficient should be very low and is more important than FRR ratio.

Revisión de algoritmos de verificación automática de firmas off-line

“…En los últimos años se han desarrollado diversos algoritmos para realizar la verificación automática de firmas offline, como redes neuronales (Kumar, Sharma y Chanda, 2012), máquina de vectores de soporte (Vargas et al, 2011), modelos ocultos de Markov (Wen, Fang, Tang y Zhang, 2009), métodos difusos (Kudłacik y Porwik, 2014), métodos basados en la distancia (Rico e Iñesta, 2012), entre otros. Asimismo, se han llevado a cabo trabajos 2.…”

The k-NN classifier and self-adaptive Hotelling data reduction technique in handwritten signatures recognition

“…Unfortunately, the repeatability of a signature features is often low and, in extreme cases, even very low [5,13,16]. For this reason, classical similarity measures applied directly to signature analysis often returns a low signature recognition level.…”

“…Recognition methods of handwritten signatures have been studied extensively and developed over many years [1-4, 7, 9-12]. Unfortunately, a reliable comparison of the different approaches is difficult due to inconsistencies in the standards applied in this field [13,14]. In practice, different databases are used, each holding a different number of original and forged signatures.…”

“…For this reason, classical similarity measures applied directly to signature analysis often returns a low signature recognition level. The selection and reduction methods are well known and frequently used in practice for the solution of various verification problems [12,13,18,19,[21][22][23]. Dimensionality reduction has been studied widely by researchers within many different scientific areas including biology, medicine, computer vision and machine learning [24][25][26][27][28][29][30].…”

“…Currently, there are many measures that can be used to specify the extent of similarity between different objects [13,[17][18][19][20], all of which are based on an analysis of the objects features. Unfortunately, the repeatability of a signature features is often low and, in extreme cases, even very low [5,13,16].…”