查看文章

Multi-biometric system is an advanced technology that has an extensive application in the area of information security. The proposed work designs a personal identification system based on a combination of biometric inputs such as finger vein and iris. In the image processing stage, Convolutional Neural Network (CNN) with different optimizers are used to steeply raise the image texture and extract high definition distinct features of the input images. The deep dream image algorithm succours CNN by strongly activating the layers of network by synthesizing the input images. This network obtains a high recognition rate, which proves to be better performing than traditional algorithms. An alert system has been adjoined with the biometric system for warning about the security breach. By this, the entire system was found efficient with better accuracy and stability has been obtained.