论文标题
使用DCGAN和Deep CNN减轻演示攻击
Mitigating Presentation Attack using DCGAN and Deep CNN
论文作者
论文摘要
基于生物识别的身份验证目前在常规身份验证系统中起着重要的作用。但是，演示攻击的风险随后增加。我们的研究旨在确定即使有足够的用户的生物特征图像样本有限，也可以防止表现攻击的领域。我们的工作着重于通过实现深卷积生成对抗网（DCGAN）来从真实图像集中产生感性的合成图像。我们已经在虚假图像生成期间实施了时间和空间增强。我们的工作使用我们的深CNN检测到对面部和虹膜图像的攻击，灵感来自VGGNET [1]。我们在三个不同的生物识别图像数据集上应用了深神经网技术，即Miche I [2]，Visob [3]和Ubipr [4]。在这项研究中使用的数据集包含在受控和不受控制的环境中捕获的图像以及不同的分辨率和大小。我们在UBI-PR [4] IRIS数据集上获得了97％的最佳测试精度。对于Miche-I [2]和Visob [3]数据集，我们分别达到了95％和96％的测试精度。
Biometric based authentication is currently playing an essential role over conventional authentication system; however, the risk of presentation attacks subsequently rising. Our research aims at identifying the areas where presentation attack can be prevented even though adequate biometric image samples of users are limited. Our work focusses on generating photorealistic synthetic images from the real image sets by implementing Deep Convolution Generative Adversarial Net (DCGAN). We have implemented the temporal and spatial augmentation during the fake image generation. Our work detects the presentation attacks on facial and iris images using our deep CNN, inspired by VGGNet [1]. We applied the deep neural net techniques on three different biometric image datasets, namely MICHE I [2], VISOB [3], and UBIPr [4]. The datasets, used in this research, contain images that are captured both in controlled and uncontrolled environment along with different resolutions and sizes. We obtained the best test accuracy of 97% on UBI-Pr [4] Iris datasets. For MICHE-I [2] and VISOB [3] datasets, we achieved the test accuracies of 95% and 96% respectively.
