Social Ski Driver-Jaya Optimization-Enabled Deep Convolution Neural Network for Signature Verification

Main Article Content

N. Neelima

Abstract

With the rapid growth of technology, security plays a very important role for avoiding forgeries and fake. It is also one of the most easily forgeable biometric identity when compared to other biometric features like thumb impression, face recognition etc.  Thus, the signature verification is the most important features to check the person authenticity.  Therefore, an effective method named Social ski driver-Jaya (SSD-Jaya) optimization algorithm is proposed in this research to verify the signature. The pre-processing is initially done by the input image in which the cropping and the binarization is done. Here, the cropping is performed for resizing the input image with the constant aspect ratio, whereas the binarization is carried out using otsu thresholding for converting the original image into binary image. After that, the feature extraction is carried out based on hierarchical skeleton, yields the addition data for enhancing the accuracy of image matching and it does not require manual intervention. The hierarchical skeleton image is fed to the formation of the graph that indicates the highly significant points in the image. At last, the classification is performed using the developed SSD-Jaya optimization algorithm, which is designed by integrating Social ski driver (SSD) and Jaya algorithm. The performance of the signature verification Proposed models are analyzed depending on the three critical measures: precision, sensitivity, and specificity. the developed model delivers the highest precision of 0.2 and sensitivity of 0.98 with respect to standard deviation

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

Article Details

How to Cite
N. Neelima. (2021). Social Ski Driver-Jaya Optimization-Enabled Deep Convolution Neural Network for Signature Verification. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(12), 62–77. https://doi.org/10.17762/turcomat.v12i12.7260
Section
Research Articles