In this paper, we have proposed a new joint architecture using Deep Neural Network (DNN) and a traditional descriptor for feature extraction towards signature identification. The proposed approach is an extended version of ResNet-18, which is enhanced using our two paths architecture. In the first path, we explore features using a deep convolutional neural network, and in the second path, we discover global features using a traditional heuristic approach. Our traditional approach extracts global features that are stable with rotation and scaling. For evaluation, we performed extensive experiments on accessible datasets of CEDAR, UTsig, and GPDS through the proposed approach. Our results show that the proposed joint approach outperformed the baseline ResNet-18 and demonstrate our approach superiority. Also, the comparisons with the related works show that our approach results are better or in par with state of the art.
Jampour, M., & Javidi, M. (2021). A Joint DNN Architecture with explicit features for Signature Identification image. Journal of Machine Vision and Image Processing, 7(2), 57-69.
MLA
Mahdi Jampour; Malihe Javidi. "A Joint DNN Architecture with explicit features for Signature Identification image". Journal of Machine Vision and Image Processing, 7, 2, 2021, 57-69.
HARVARD
Jampour, M., Javidi, M. (2021). 'A Joint DNN Architecture with explicit features for Signature Identification image', Journal of Machine Vision and Image Processing, 7(2), pp. 57-69.
VANCOUVER
Jampour, M., Javidi, M. A Joint DNN Architecture with explicit features for Signature Identification image. Journal of Machine Vision and Image Processing, 2021; 7(2): 57-69.