Fingerprint is one of the most widely explored biometric traits. Specifically, contact-based fingerprint recognition systems reign supreme due to their effectiveness, portability, and research advancements in this field. However, these are issues such as hygiene concerns, sensor degradation due to constant physical contact, and latent fingerprint threats. In this paper, we present the details of a fingerprint recognition system that captures finger photos without contact. The captured finger photos are then processed to obtain global and local (minutiae-based) features. Specifically, a Siamese convolutional neural network is designed to extract global features from a given finger photo. Finally, the matching scores computed using the global and local features are fused to generate the final matching score between the probe and the reference fingerprint templates. We have developed the proposed system using the Nvidia Jetson Nano development kit. The developed system performs contactless fingerprint recognition in real time with minimum latency and acceptable matching accuracy. The performance of the proposed system is evaluated on an in-house IITI contactless fingerprint dataset (IITI-CFD) containing 105 train and 100 test subjects. The proposed system achieves an equal error rate of 2.19% on IITI-CFD.