Recognition of Dorsal Hand Vein in Small-Scale Sample Database Based on Fusion of ResNet and HOG Feature

Abstract

As artificial intelligence develops, deep learning algorithms are increasingly being used in the field of dorsal hand vein (DHV) recognition. However, deep learning has high requirements regarding the number of samples, and current DHV datasets have few images. To solve the above problems, we propose a method based on the fusion of ResNet and Histograms of Oriented Gradients (HOG) features, in which the shallow semantic information extracted by primary convolution and HOG features are fed into the residual structure of ResNet for full fusion and, finally, classification. By adding Gaussian noise, the North China University of Technology dataset, the Shandong University of Science and Technology dataset, and the Eastern Mediterranean University dataset are extended and fused to from a fused dataset. Our proposed method is applied to the above datasets, and the experimental results show that our proposed method achieves good recognition rates on each of the datasets. Importantly, we achieved a 93.47% recognition rate on the fused dataset, which was 2.31% and 26.08% higher than using ResNet and HOG alone.

Backpropagation Design for Authenticating Blood Vessel Patterns of the Back of the Hand Using GLRLM

“…Dalam konteks keamanan digital, pengenalan biometrik menawarkan solusi yang sangat efektif karena memanfaatkan karakteristik unik individu sebagai kunci identifikasi. Pembuluh darah punggung tangan menawarkan kelebihan tertentu yang menjadikannya pilihan menarik untuk sistem pengenalan biometrik karena pola pembuluh darah punggung tangan cenderung stabil sepanjang waktu, tidak dipengaruhi oleh faktor eksternal seperti perubahan cuaca atau kebersihan [13]. Citra pembuluh darah mempunyai sifat unik karena sulit dipalsukan, diubah, atau dirusak karena letaknya di kulit manusia.…”

“…Penelitian sebelumnya juga dilakukan pada tahun 2022 dengan judul "Recognition of Dorsal Hand Vein in Small-Scale Sample Database Based on Fusion of ResNet and HOG Feature" [13]. Pada penelitian ini melakukan pengenalan pembuluh darah punggung tangan terhadap 372 individu dan setiap individu dilakukan pengambilan citra sebanyak 2 kali, pra pengolahan citra dilakukan dengan metode RGB to Citra Keabuan, Resize, Filter Median, dan Gaussian, ekstraksi ciri yang digunakan yaitu Histograms of Oriented Gradients (HOG) dan ResNet.…”

Dorsal Hand Vein Image Recognition: A Review

“…A transfer learning approach is used to adjust the CNNs' final output layers to correspond to the number of classes contained in the datasets.  Kefeng Li et al [40] presented an approach based on the fusion of ResNet and Histograms of Oriented Gradients (HOG) features, in which the residual structure of ResNet is fed with the shallow semantic data generated by primary convolution and HOG features for full fusion and, ultimately, classification is done based on using KNN classifier.…”