Biometric systems proven to be one of the most reliable and robust method for human identification. Integration of biometrics among the standard of living provokes the necessity to vogue secure authentication systems. The use of palm-prints for user access and authentication has increased in the last decade. To give the essential security and protection benefits, conventional neural networks (CNNs) has been bestowed during this work. The combined CNN and feature transform structure is employed for mapping palm-prints to random base-n codes. Further, secure hash algorithm (SHA-3) is used to generate secure palm-print templates. The proficiency of the proposed approach has been tested on PolyU, CASIA and IIT-Delhi palm-print datasets. The best recognition performance in terms of Equal Error Rate (EER) of 0.62% and Genuine Acceptance Rate (GAR) of 99.05% was achieved on PolyU database.