REGULAR paper has been Published by IEEE T-CSVT (IF: 3.558)
发布时间：2019-03-12
点击次数：
- 发布时间：
- 2019-03-12
- 文章标题：
- REGULAR paper has been Published by IEEE T-CSVT (IF: 3.558)
- 内容：
Centralized Large Margin Cosine Loss for Open-set Deep Palmprint Recognition
As one promising branch of biometrics, palmprint recognition has received significant attention and made extraordinary progress in the past decades. The crucial step of palmprint recognition is to extract discriminative features for the subsequent identification or verification task. However, neither the traditional hand-crafted descriptors nor deep convolutional neural network (CNN) with the original softmax loss show satisfactory generalization ability under open-set settings. In this paper, we proposed an end-to-end method for open-set palmprint recognition by applying CNN with a novel loss function, namely centralized large margin cosine loss (C-LMCL). The modified loss function compels feature vectors from different classes to uniformly and separately distribute in the hyper feature space. At the same time, it makes intra-class feature vectors compactly gather to their corresponding class centers. Consequently, such trained model has the ability to generalize across unseen subjects and different datasets. Finally, a lot of experiments are conducted on two public palmprint datasets – Tongji and PolyU datasets. In particular, all the evaluation are made under open-set protocols, which are more complex and challenging compared to previous close-set scenarios. The experimental results on the Tongji and PolyU datastes indicate the superiority of our algorithm over the state-of-the-art performance. It effectively confirmed the bright prospects of employing palmprint information in biometric authentication.