In computer science considering the large scale network and increasing number of application, the protection and security of the system is essential. The Information Technology System is improving day by day. The growing concern of computer science technology creates problem for identity theft problems, hence protection and computer system security is playing vital role. The conventional alphanumeric password protection system is very vulnerable. This kind of alphanumeric password can be traced and difficult to remember for the system user. This leads to a new era of protection and security system in case of authenticity in computer science. In this paper, it is proposed to have very user-friendly secure multilevel biometric graphical authentication system. This kind of authentication system is easy for authorized user but very difficult to un- authorized user to trace. The multilevel and biometric-graphical secure authentication system algorithm described here uses more than one level of authentication system such as face detection, DNA finger printing, image pattern matching and alphanumeric password with image pattern. This algorithm based on entities that are actually bound with the individual at a much secure level than. As a result, they are more reliable since biometric information cannot be lost, forgotten, or guessed easily. Biometrics system is being used in the identification of individuals based on some physical attribute. The physical attributes may be the finger print, face detection, retina scanning and voice detection. This system is widely used in authentication system. The traditional system of alphanumeric password and PIN no protection system in magnetic flip can be damaged and can be read by decoding the flip and can be forgotten and disclosed any time. The biometric technology provides a strong user-friendly authentication solution. The combination of biometric and other protocols can provide a sophisticated highly secure system. Biometrics technology more efficient way of authentication than the more common use of passwords, smart cards, or a combination of the two. The system user need not required to remember the series of passwords. Normal alphanumeric password expires after certain interval of time and need to assign again hence keeping tract is also difficult. Software industries, research institutes, banking organization where the protection and security of the system is extensively experiences for the sensitive data and applications and for all kind of application and systems there may be series of passwords. As a result for ease of use with high level user friendly security system can only be the solution in such case. The biometric face detection and pattern matching technique is very latest and effective technique to provide high level security. The pattern matching technique is convenient to use in this system of authentication and is very unique. In case of face detection algorithm human face with all its facial appearances including node point, eye, forehead, cheeks etc i.e identification of facial points (nose, eyes, checks, lips, ears