-Signature identification and verification are of great importance in authentication systems. The purpose of this paper is to introduce an experimental contribution in the direction of multi-script off-line signature identification and verification using a novel technique involving off-line English, Hindi (Devnagari) and Bangla (Bengali) signatures. In the first evaluation stage of the proposed signature verification technique, the performance of a multi-script off-line signature verification system, considering a joint dataset of English, Hindi and Bangla signatures, was investigated. In the second stage of experimentation, multi-script signatures were identified based on the script type, and subsequently the verification task was explored separately for English, Hindi and Bangla signatures based on the identified script result. The gradient and chain code features were employed, and Support Vector Machines (SVMs) along with the Modified Quadratic Discriminate Function (MQDF) were considered in this scheme. From the experimental results achieved, it is noted that the verification accuracy obtained in the second stage of experiments (where a signature script identification method was introduced) is better than the verification accuracy produced following the first stage of experiments. Experimental results indicated that an average error rate of 20.80% and 16.40% were obtained for two different types of verification experiments.
INTRODUCTION
Biometrics are the most widely used approaches for
personal identification and verification. Among all of the
biometric authentication systems, handwritten signatures, a
pure behavioral biometric, have been accepted as an official
means to verify personal identity for legal purposes on such
documents as cheques, credit cards and wills [
Considerable research has previously been undertaken in the area of signature verification, particularly involving single-script signatures. On the other hand, less attention has been devoted to the task of multi-script signature verification. Very few published papers involving multiscript signatures, including non-English signatures, have been communicated in the field of signature verification.
Pal et al. [
Development of a general multi-script signature
verification system, which can verify signatures of all
scripts, is very complicated. The verification accuracy in
such multi-script signature environments will not be as
successful when compared to single script signature
verification [
The remainder of this paper is organized as follows. The multi-script signature verification concept is described in Section II. Section III introduces the notable properties of Hindi and Bangla script. The Hindi, Bangla and English signature database used for the current research is described in Section IV. Section V briefly presents the feature extraction techniques employed in this work. The classifier details are described in Section VI. The experimental settings are presented in Section VII. Results and a discussion are provided in Section VIII. Finally, conclusions and future work are discussed in Section IX.
II.
When a country deals with two or more scripts and languages for reading and writing purposes, it is known as a multi-script and multi-lingual country. In India, there are officially 23 (Indian constitution accepted) languages and 11 different scripts.
In such a multi-script and multi-lingual country like India, languages are not only used for writing/reading purposes but also applied for reasons pertaining to signing and signatures. In such an environment in India, the signatures of an individual with more than one language (regional language and international language) are essentially needed in official transactions (e.g. in passport application forms, examination question papers, money order forms, bank account application forms etc.). To deal with these situations, signature verification techniques employing single-script signatures are not sufficient for consideration. Therefore in a multi-script and multi-lingual scenario, signature verification methods considering more than one script are necessarily required.
Towards this direction of verification, the contribution of this paper is twofold: First, multi-script signature verification considering joint datasets as shown in Figure 1, the second is identification of signatures based on script, and subsequent verification for English, Hindi and Bangla signatures based on the identified script result. A diagram of this second verification mode is shown in Figure 2.
Multi-script off-line Signatures (Signatures
of English, Hindi and Bangla)
Verification based on Multi-script Signatures
Most of the Indian scripts including Bangla and Devanagari
have originated from ancient Brahmi script through various
transformations and evolution [
IV.
As there has been no public signature corpus available for Hindi and Bangla script, it was necessary to create a database of Hindi and Bangla signatures. The Hindi and Bangla signature databases used for experimentation consisted of 50 sets per script type. Each set consists of 24 genuine signatures and 30 skilled forgeries. Some genuine signature samples of Hindi and Bangla, with their corresponding forgeries, are displayed in Table 1 and Table 2.
Another database, consisting of 50 sets from GPDS-160 [
Feature extraction is a crucial step in any pattern recognition system. Two different types of feature extraction techniques such as: gradient feature extraction and the chain code feature are considered here.
576-dimensional gradient features were extracted for this
research and experimentation, which are described in paper
[
The 64-dimensional Chain Code feature is determined as follows. In order to compute the contour points of a twotone image, a 3 x 3 window is considered surrounding the object point. If any one of the four neighbouring points (as shown in Fig. 3 (a)) is a background point, then this object point (P) is considered as a contour point. Otherwise it is a non-contour point.
The bounding box (minimum rectangle containing the character) of an input character is then divided into 7 x 7 blocks. In each of these blocks, the direction chain code for each contour point is noted and the frequency of the direction codes is computed. Here, the chain code of four directions only [directions 1 (horizontal), 2 (45 degree slanted), 3 (vertical) and 4 (135 degree slanted)] is used. Four chain code directions are shown in Fig. 3 (b). It is assumed that the chain code of directions 1 and 5, 2 and 6, 3 and 7, 4 and 8, are the same. Thus, in each block, an array is obtained of four integer values representing the frequencies, and those frequency values are used as features. Thus, for 7 x 7 blocks, 7 x 7 x 4= 196 features are obtained. To reduce the feature dimensions, after the histogram calculation into 7 x 7 blocks, the blocks are down-sampled with a Gaussian filter into 4 x 4 blocks. As a result, 4 x 4 x 4 = 64 features are obtained for recognition. To normalize the features, a maximum value of the histograms from all the blocks, is computed. Each of the above features is divided by this maximum value to obtain the feature values between 0 and 1.
(a) (b) Figure 3. Eight neighbours (a) For a point P and its neighbours (b) For a point P the direction codes for its eight neighbouring points.
VI.
CLASSIFIER DETAILS
Based on these features, Support Vector Machines (SVMs) and the Modified Quadratic Discriminant Function (MQDF) are applied as the classifiers for the experiments.
For this experiment, a Support Vector Machine (SVM) classifier is used. The SVM is originally defined for twoclass problems and it looks for the optimal hyper plane, which maximizes the distance and the margin, between the nearest examples of both classes, named support vectors (SVs). Given a training database of M data: {xm| m=1,..., M}, the linear SVM classifier is then defined as: f (x) j x j x b
j
where {xj} are the set of support vectors and the parameters
j and b have been determined by solving a quadratic
problem [
The Modified Quadratic Discriminant Function is defined as
follows [
( ) ( where X is the feature vector of an input character; M is a mean vector of samples; is the ith eigen vector of the sample covariance matrix; is the ith eigen value of the sample covariance matrix; k is the number of eigen values considered here; n is the feature size; is the initial estimation of a variance; N is the number of learning samples; and N0 is a confidence constant for s and N0 is considered as 3N/7 for experimentation. All the eigen values and their respective eigen vectors are not used for classification. Here, the eigen values are stored in descending order and the first 60 (k=60) eigen values and their respective eigen vectors are used for classification. Compromising on trade-off between accuracy and computation time, k was determined as 60.
VII.
A. Settings for Verification used in 1st Stage of Experiments The skilled forgeries were not considered for training purposes. For experimentation, random signatures were considered for training purposes. For each signature set, an SVM was trained with 12 randomly chosen genuine signatures. The negative samples for training (random signatures) were the genuine signatures of 149 other signature sets. Two signatures were taken from each set. In total, there were 149x2=298 random signatures employed for training. For testing, the remaining 12 genuine signatures and 30 skilled forgeries of the signature set being considered were employed. The number of samples for training and testing for these experiments are shown in Table 3.
Table 3. No. of Signatures used per set in 1st Phase of Verification Training
Genuine Signature 12 1) Settings for Signature Script Identification 150 sets of signatures (50 sets of English, 50 sets of Hindi and 50 sets of Bangla) were used for signature script identification. 30 sets of signatures from each script were considered for training, and the remaining 20 sets were considered for testing purposes. The number of samples for training and testing used in experimentation of the identification approach are shown in Table 4.
The verification task in the second stage was explored separately for English signatures, Hindi signatures and Bangla signatures based on the identified script result. Signature samples (30 sets from each script) that were considered for training purposes in signature script identification were not used for the individual verification task. Only the correctly identified samples from 20 sets (used for the testing part in identification) were considered for verification. For each signature set, an SVM was trained with 12 genuine signatures. The negative samples for training were 95 (19x5) genuine signatures of 19 other signature sets.
In this stage of experimentation, 8100 (150x54) signatures involving English, Hindi and Bangla scripts were employed for training and testing purposes. At this operational point, the SVMs produced an AER of 20.80%, and an encouraging accuracy of 79.20% was achieved in this first mode of verification.
In this stage of verification the signatures are identified based on their script and subsequently, the identified signatures are applied separately for verification. In the signature script identification stage, only 64-dimensional chain code features were used because a slightly better accuracy was obtained when compared to the gradient feature. The MQDF classifier was also taken into account in the script identification step applying chain code features for a better accuracy, but MQDF did not achieve the better result as compared to SVMs in this study. To get a comparative idea, script identification results using two different classifiers with chain code features are shown in Table 5. An accuracy of 93.08% is achieved at the script identification stage by using the SVM classifier. The accuracy of Bangla, English and Hindi are 85.19, 95.74 and 98.33% respectively. Confusion matrices obtained using SVM classifiers, and the 64-dimensional chain code features investigated, are shown in Table 6. Based on the outcomes of the identification phase, verification experiments subsequently followed. Verification results obtained for individual scripts were calculated on 93.08% (identification rate) accuracy levels. In this phase of experimentation, the SVMs produced an overall AER of 21.10%, 13.05% and 15.05% using English, Hindi and Bangla signatures respectively. The overall verification accuracy obtained for the second major experiments (identification plus verification) was 83.60% (average of 78.90% of English, 86.95% of Hindi and 84.94% of Bangla).
From the experimental results obtained, it was observed that the performance of signature verification in the second set of experiments (identification and verification) was encouraging compared to the signature verification accuracy from the first experiment set (verification only). Table 7 demonstrates the accuracies attained in the first experiment set as well as separate verification results for English, Hindi and Bangla from the second experiment set. In the second stage of verification, the overall accuracy is 83.60% (Avg. of 78.90%, 86.95% and 84.94%) which is 4.40 (83.60-79.20) higher than the accuracy in the first stage. The comparison of these two accuracies is shown in Table. 8. From the above table it is evident that verification accuracy with script identification is much higher than without script identification. This increased accuracy is achieved because of the proper application of the identification stage. This research clearly demonstrates the importance of using identification in multi-script signature verification techniques.
Most of the methods used for signature verification generate some erroneous results. In these experiments, a few signature samples were mis-identified in both the identification and verification stages. Few of the confusing signature samples obtained in the signature script identification stage using the SVM classifier are shown in Figures 4, 5 and 6. Three categories of confusing samples are generated by the classifier. The first category illustrates a Bangla signature sample treated as a Hindi signature sample. The second one illustrates an English signature sample treated as a Bangla signature sample and the third one illustrates a Hindi signature sample treated as a Bangla signature sample.
This paper provides an investigation of the excellent performance of a multi-script signature verification technique involving English, Hindi and Bangla off-line signatures. The novel approach used in a multi-script signature verification environment with the combination of a custom Hindi and Bangla off-line signature dataset provides a substantial contribution to the field of signature verification. In such a verification environment, the proper utilization of a script identification technique, which substantially affects the verification accuracy, indicates an important step in the process. The comparatively higher verification accuracy obtained in the second experimental approach is likewise a substantial contribution. The gradient feature, chain code feature as well as SVM and MQDF classifiers were employed for experimentation. The idea of a multi-script signature verification approach, which deals with an identification phase, is a very important contribution to the area of signature verification. The proposed off-line multi-script signature verification scheme is a new investigation in the field of off-line signature verification. In the near future, we plan to extend our work considering further sets of signature samples, which may include different languages/scripts.
Thanks to my colleague Mr. Nabin Sharma for his help towards the preparation of this paper.