Fingerphoto images captured using a smartphone are successfully used to verify the individuals that have enabled several applications. This work presents a novel algorithm for fingerphoto verification using a nested residual block: Finger-NestNet. The proposed Finger-NestNet architecture is designed with three consecutive convolution blocks followed by a series of nested residual blocks to achieve reliable fingerphoto verification. This paper also presents the interpretability of the proposed method using four different visualization techniques that can shed light on the critical regions in the fingerphoto biometrics that can contribute to the reliable verification performance of the proposed method. Extensive experiments are performed on the fingerphoto dataset comprised of 196 unique fingers collected from 52 unique data subjects using an iPhone6S. Experimental results indicate the improved verification of the proposed method compared to six different existing methods with EER = 1.15%.