Fingerprint recognition based on collected images using deep learning technology

The fingerprint identification is the most widely used authentication system. uniqueness for each human being provides error-free identification. However, during scanning process of fingerprint, generated image using scanner may differ slightly scan. This paper proposes an efficient matching model deep learning based convolutional neural network (CNN or ConvNet). proposed CNN consists fifteen layers and classified into two stages. first stage preparation which includes images collection, augmentation pre-processing steps, while second features extraction stage. Regarding implantation results, system provided best given features. obtained training accuracy 100% dataset validating dataset.