**4.1 Classification based on SVM**

For offline Arabic signature verification and identification Support Vector Machines (SVM) was used. Important features in the Arabic signature images were extracted and the samples were confirmed with the assistance of Gaussian empirical law. SVM was applied to record corresponding results for comparing all signatures from database with the test signature. The suggested method is tested on Arabic signatures containing 500 samples of 50 users and the outcomes are obtained to be encouraging. In a high dimension feature area the principle of SVM, depends on a linear isolation where information were mapped to take into consideration the final non-linearity of the issue. SVM classifier [22, 23] was trained with corresponding

#### **Figure 3.**

*The different stages of the pre-processing phase, (a) gray sample, (b) binary and converted sample, (c) with boundary box sample, (d) resized image, (e) windowing image.*

*New Attributes Extraction System for Arabic Autograph as Genuine and Forged through… DOI: http://dx.doi.org/10.5772/intechopen.96561*

result vectors for each distance. This is to obtain a good level of generalization capability. To establish the rating of signers' relationship to the inquiry samples, firstly we used these processing points and then we combined the results of the entire samples.

#### **4.2 Decision tree classification**

using the EPW algorithm. These were calculated based on Eqs. (2) and (3) to

Normalization of information ensures the genuineness or forgery of signatures in the training set. We trained a decision tree classifier to recognize the genuine and forged signatures in this normalized feature area (**Figure 3**). To facilitate comparisons, two classifiers were used: The Tree classifier and SVM classifier were applying the 2-dimensional attribute vectors. A linear classification was made by choosing a threshold ratio separating the two classes within the training set. This threshold was

For offline Arabic signature verification and identification Support Vector Machines (SVM) was used. Important features in the Arabic signature images were extracted and the samples were confirmed with the assistance of Gaussian empirical law. SVM was applied to record corresponding results for comparing all signatures from database with the test signature. The suggested method is tested on Arabic signatures containing 500 samples of 50 users and the outcomes are obtained to be encouraging. In a high dimension feature area the principle of SVM, depends on a linear isolation where information were mapped to take into consideration the final non-linearity of the issue. SVM classifier [22, 23] was trained with corresponding

*The different stages of the pre-processing phase, (a) gray sample, (b) binary and converted sample, (c) with*

*boundary box sample, (d) resized image, (e) windowing image.*

N max ¼ dmax*=*Pmax (2) N min ¼ dmin*=*Pmin (3)

represent the allocation of the feature group.

used in the verification process.

*Applications of Pattern Recognition*

**4.1 Classification based on SVM**

**Figure 3.**

**100**

Evaluation of Tree Classification (Bagged Trees) technique was used in the same way and on the same samples from Arabic signatures as SVM. MATLAB 2014 bagged tree classification and trees software were used in the training and classification simulation. To predict a reaction, the decision procedure in the decision tree from the root (starting) node (feature) down to a leaf (feature) node was followed. Responses were included in the leaf feature. Decision trees granted responses, such as 'true' or 'false'. Decision Tree was created to perform classification [20, 24]. The described steps are presented in Algorithm 1.


## **5. Outcomes and discussion**

In this section, we discuss the outcomes of the suggested methodology on some of samples from the Arabic signatures.

#### **5.1 Pre-processing**

The input image in RGB color space was first converted to grayscale image as displayed in **Figure 3(a)** represented Gray image. Then, the image was smoothened with median filter and converted to binary as shown in **Figure 3(b)**. Further, the image was passed from boundary box to find the boundaries of the text area as presented in (c), while in (d) the image was resized to apply the adaptive windowing algorithm to divide it into fragments as shown in (e).

#### **5.2 Feature extraction**

In this phase, we represent the sub-images from a set of features. The outcome of the feature extraction is shown in **Table 1(a)**. Initially, these features were not normalized. The values shown in **Table 1(a)** represent the frequencies of the designs extracted from each box. Higher ratios mean there is a more specific model with the genuine autograph, which suggests that the Arabic signatures are highly similar to the test signature. The features were then normalized using a composed matrix of features. The projection and profile features were normalized using window height, while the other descriptors were normalized by their respective maximum possible value. Normalization places different feature values in the same ranges as shown in **Table 1(b)**. After normalization, each normalized feature of main window were concatenated into a single feature set, which represent each


window by a vector. This process can standardize all features by scaling each

*Feature extraction un-normalized and normalized. (a) Un-normalized features (b)Normalization.*

**F10 F9 F8 F7 F6 F5 F4 F3 F2 F1** 0.6633 0.3205 0.9257 0.1098 0.304 0.6633 0.5974 0.3205 0.9257 0.1098 0.794 0.3367 0.9165 0.0582 0.056 0.794 0.4185 0.3367 0.9165 0.0582 0.6213 0.8938 0.9129 0.4802 0.199 0.6213 0.0595 0.8938 0.9129 0.4802 **F20 F19 F18 F17 F16 F15 F14 F13 F12 F11** 0.299 0.611 0.547 0.725 0.581 0.910 0.299 0.611 0.725 0.581 0.6727 0.741 0.0484 0.62 0.76 0.0157 0.6727 0.741 0.62 0.76 0.386 0.705 0.348 0.2794 0.472 0.523 0.386 0.705 0.2794 0.472 0.8651 0.925 0.6883 0.6446 0.352 0.0523 0.8651 0.925 0.6446 0.352 0.952 0.274 0.964 0.424 0.398 0.1568 0.952 0.274 0.424 0.398 0.4175 0.645 0.2759 0.612 0.152 0.6021 0.4175 0.645 0.612 0.152 0.915 0.722 0.7266 0.6831 0.376 0.2531 0.915 0.722 0.6831 0.376 0.9235 0.596 0.794 0.1576 0.424 0.3471 0.9235 0.596 0.1576 0.424 0.4185 0.666 0.9817 0.021 0.411 0.6649 0.4185 0.666 0.021 0.411 0.1315 0.231 0.9571 0.4585 0.024 0.8789 0.1315 0.231 0.4585 0.024 0.3969 0.666 0.4075 0.1098 0.304 0.6673 0.3969 0.666 0.1098 0.304 0.2144 0.754 0.8988 0.0582 0.056 0.7794 0.2144 0.754 0.0582 0.056 0.479 0.461 0.016 0.482 0.199 0.6613 0.479 0.461 0.482 0.199

*New Attributes Extraction System for Arabic Autograph as Genuine and Forged through…*

When the procedure of feature selection technique for windows was accomplished, those features with sufficient number of windows were kept. The features contained stroke patterns occurring in the windows. Generally, the number of patterns for each feature selection was proportional to the size of the Arabic signature sample. According to **Figure 4**, one important point to note is the number of selected features. This is a property of the signer as can be observed from **Figure 5**, where the number of selected features are presented. In this case, feature selection

feature to a given range.

**Table 1.**

**Figure 4.**

**103**

*After Selection Feature step by SPCA method.*

**(b) Normalization**

*DOI: http://dx.doi.org/10.5772/intechopen.96561*

**5.3 Representation of feature selection**

*New Attributes Extraction System for Arabic Autograph as Genuine and Forged through… DOI: http://dx.doi.org/10.5772/intechopen.96561*


#### **Table 1.**

**(a) Un-normalized features**

*Applications of Pattern Recognition*

**(b) Normalization**

**102**

**F10 F9 F8 F7 F6 F5 F4 F3 F2 F1** 3.000 2.0020 3.0000 1.0001 3.000 1.0000 1.0000 6.0000 1.0001 3.000 1.0000 3.0000 1.0000 1.0000 1.0000 1.0000 1.0000 8.0000 1.0000 1.0000 1.0000 4.0000 1.0000 1.0000 1.0000 1.0000 1.0000 9.0000 1.0000 1.0000 1.0000 3.0000 1.0000 1.0000 1.0000 1.0000 1.0000 11.0000 1.0000 1.0000 1.0000 6.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.20000 1.0000 1.0000 1.0000 8.0000 1.0000 2.0000 1.0000 2.0000 2.0000 10.0000 2.0000 1.0000 2.0000 2.6463 2.0000 3.0000 2.0000 2.6463 2.6463 10.0000 3.0000 2.0000 3.0000 3.9281 3.0000 1.0000 3.0000 3.9281 3.9281 9.0000 1.0000 3.0000 4.0000 2.491 4.0000 1.0000 4.0000 2.491 2.491 6.0000 1.0000 4.0000 3.0000 1.8671 3.0000 1.0000 3.0000 1.8671 1.8671 3.0000 1.0000 3.0000 6.0000 1.3205 6.0000 1.0000 6.0000 1.3205 1.3205 1.0000 1.0000 6.0000 8.0000 2.6463 8.0000 1.0000 8.0000 0.3367 0.3367 1.0000 1.0000 8.0000 1.0000 0.9079 9.0080 1.3205 1.0000 0.836 0.839 1.0000 1.0000 9.0900 **F20 F19 F18 F17 F16 F15 F14 F13 F12 F11** 3.0000 2.0020 3.0000 4.088 2.6106 3.0000 1.0001 3.000 1.0000 6.0000 1.0000 3.0000 1.0000 4.764 1.057 3.0000 1.0000 1.0000 1.0000 8.0000 1.0000 4.0000 1.0000 4.472 1.5523 3.0000 1.0000 1.0000 1.0000 9.0000 1.0000 3.0000 1.0000 7.352 0.0523 1.0000 1.0000 1.0000 1.0000 11.0000 1.0000 6.0000 1.0000 5.336 0.1469 1.0000 1.0000 1.0000 1.0000 1.20000 1.0000 8.0000 1.0000 3.152 1.6021 2.9066 2.0000 1.0000 2.0000 10.0000 2.0000 2.6463 2.0000 3.376 1.0000 1.6974 3.0000 2.0000 2.6463 10.0000 3.0000 3.9281 3.0000 6.424 1.0000 1.0000 1.0000 3.0000 3.9281 9.0000 4.0000 2.491 4.0000 2.4 1.0000 5.0000 1.0000 4.0000 2.491 6.0000 3.0000 1.8671 3.0000 0.024 1.0000 4.0000 1.0000 3.0000 1.8671 3.0000 6.0000 1.3205 6.0000 0.304 1.0000 5.0000 1.0000 6.0000 1.3205 1.0000 8.0000 2.6463 8.0000 0.056 2.0000 6.0000 1.0000 8.0000 0.3367 1.0000 9.0080 1.0200 3.0010 0.464 1.3205 1.0000 1.3205 1.0000 0.836 1.0200

**F10 F9 F8 F7 F6 F5 F4 F3 F2 F1** 0.690 0.57 0.824 0.635 0.041 0.61 0.68 0.72 0.83 0.65 0.087 0.345 0.760 0.622 0.706 0.157 0.156 0.395 0.760 0.67 0.523 0.875 0.8887 0.2794 0.472 0.523 0.0921 0.875 0.8887 0.2794 0.0523 0.477 0.3109 0.6446 0.352 0.0523 0.4585 0.477 0.3109 0.6446 0.1468 0.322 0.5577 0.424 0.396 0.1468 0.1098 0.322 0.5577 0.424 0.6021 0.8581 0.9066 0.6012 0.152 0.6021 0.0582 0.8581 0.9066 0.6012 0.2531 0.6463 0.6974 0.6831 0.376 0.2531 0.4802 0.6463 0.6974 0.6831 0.3451 0.9281 0.7784 0.1576 0.424 0.3451 0.2093 0.9281 0.7784 0.1576 0.6649 0.491 0.9262 0.0621 0.411 0.6649 0.6716 0.491 0.9262 0.0621 0.8189 0.8671 0.9862 0.4585 0.024 0.8189 0.1161 0.8671 0.9862 0.4585

*Feature extraction un-normalized and normalized. (a) Un-normalized features (b)Normalization.*

window by a vector. This process can standardize all features by scaling each feature to a given range.

## **5.3 Representation of feature selection**

When the procedure of feature selection technique for windows was accomplished, those features with sufficient number of windows were kept. The features contained stroke patterns occurring in the windows. Generally, the number of patterns for each feature selection was proportional to the size of the Arabic signature sample. According to **Figure 4**, one important point to note is the number of selected features. This is a property of the signer as can be observed from **Figure 5**, where the number of selected features are presented. In this case, feature selection

**Figure 4.** *After Selection Feature step by SPCA method.*

features that remain consistent and stable despite the interference of influencing factors. Such natural features are almost impossible to imitate, even by the original

*New Attributes Extraction System for Arabic Autograph as Genuine and Forged through…*

The comparison between Arabic signature recognition methods were by verification rate and not by the computational time. The accuracy performance measure has been computed using confusion matrix Where, TP signifies the number of true positive signatures, TN refers to the number of true negative signatures, FP signifies the number of false positive cases and FN signifies the number of false negative signatures. True Positive Ratio is the measure of genuine signatures classified correctly as genuine; False Positive Rate is the measure of a forgery signature classified as genuine. False Negative Rate is the measure of a genuine situation classified as forgery. True Negative Proportion is a measure of a forgery signature classified as

The Verification Rate ¼ ðTP þ TNÞ*=*ðTP þ TN þ FP þ FNÞ � 100%

**Authors Methods Language Verification Rate**

Arabic 98%

by fuzzy concepts

A.y. Ebrahim [26] DCT+ DWT Technique Arabic 99.75% SM Darwish, al., [27] Distance and Fuzzy Classifiers Alliance Arabic 98% C. Ergun al., [28] word layout signature Farsi 94:3% Proposed method (2021) DCT+SPCA features Technique Arabic 99.8%

*An assessment table relating between the projected Arabic signature recognition system based on Arabic*

The tests assumed that 99.8% accuracy proportion Predicted Valu and Decision Tree. Such promising results are pinpointing the state-of-the-art preprocessing techniques and best performance of proposed features to discriminate between genuine and forged signatures with higher accuracy rate. The authentication of the achievements of the suggested method was achieved applying the verification rate using DCT + SPCA method were computed and compared against the two other vastly agreeable autograph verification methods. **Table 3** shows the simulation results with the Arabic signatures consisting 2000 signatures from 100 various signers. The validation rate for the proposed technique is 99.8% attesting to its superiority against the others. We could conclude that DCT + SPCA features technique and Decision Tree classifier was a credible and reliable technique for

This paper, we described a method we developed to important features selection using DCT + SPCA features technique in offline Arabic signature verification. It employed the partition of signature samples into 14x14 windows and generated the features extracted for each window. Then, this feature selection was used for

signers.

forgery.

**7. Conclusion**

**Table 3.**

**105**

classification techniques.

**6. Authentication of results**

*DOI: http://dx.doi.org/10.5772/intechopen.96561*

verification of offline Arabic signatures.

Ismail al. [25] New procedures for autograph verification

*signatures and other signatures with other previously known approaches.*

#### **Figure 5.**

*The number of selected important features of DCT + SPCA method for the two samples of 40 signers.*

is generated from 40 different signers using two tasters from each one. As can be realized, the bows represent the number of selected features in the two tasters of the same writer are close to each other for DCT + SPCA method. This seems consistent with the supposition that the value of selected features is a signer-dependent feature.

#### **5.4 Matching**

The matching phase is when the model is created using Classification and Regression Tree (Tree) and Support Vector Machines Classification (SVM) with different input parameters. Based on a person's signature, a model was created for the original and forgery signatures. The performance of the proposed method on 100 signers from Arabic signatures were used in identification for classification using DCT + SPCA features for selected important features with SVM classifier achieved the verification rate of 98.7%, and EER of 1.90% and same DCT + SPCA features with Tree classification achieved the verification rate of 99.8%, and the EER dropped to 1.20%. which was better than other techniques, as shown in **Table 2**. The objective of this study was creating a system that 1) can identify handwritten signatures and verify their authenticity, and 2 distinguish forgery from genuine ones, and those created under pressure and other influences. Using 2000 Arabic signatures samples. The results of the matching phase are shown in **Table 2**.

This implies that a forger may not skillfully repeat all aspects of the original signature. It also shows a pattern in forgers, which has small variations. Evidence shows that the mean of a feature produced by a forger in multiple attempts at forging tends to lie in a small range. Conversely, genuine signatures produced by a signer may vary under unusual conditions. Signers possess certain unconscious


**Table 2.**

*Experimental results obtained from 100 signer based on Arabic signatures.*

*New Attributes Extraction System for Arabic Autograph as Genuine and Forged through… DOI: http://dx.doi.org/10.5772/intechopen.96561*

features that remain consistent and stable despite the interference of influencing factors. Such natural features are almost impossible to imitate, even by the original signers.
