This report investigates an automated fingerprinting system for the analysis and recognition of fingerprint image. It looks at ways to enhance the performance of the Artificial Neutral Network, so that reliability in detecting minutiae is achieved in optimal time. With the increased use of computers as vehicles of information technology, it is easier for an individual to gain restrict access to sensitive/personal data by replacing PINs or passwords. Biometric techniques can potentially prevent unauthorized access to or fraudulent use of all financial and other kinds of systems holding confidential information. Fingerprinting is one of the oldest biometrics technologies. Due to the cost and usage of these kinds of systems, they are being deployed in a large number of civilian applications. For any pattern recognition system it’s relatively important to extract the important features of an image which in our case is of a fingerprint. So, an efficient and reliable mechanism for extracting these important features called minutiae is strongly needed. This report proposes a method that fulfills the needs above by using an artificial neural network. This technique is used due to the resistance it offers against noise, the promising results it offers and its ability to make hard tasks easier for computers by making up a complex system easier to handle through adapting the characteristics of biological neurons. In every image recognition system, it is often a practice to perform preprocessing before analyzing the image. Our proposed method is capable of handle all sorts of discrepancies that occur in fingerprint image. Therefore, it allows us to skip preprocessing computation and saves up time by working on the gray image straight away. Our method outputs three distinct features found in fingerprint, therefore it provides more data for authentication and verification of an individual. This feature of our method offers a lot of reliability in matching process.