Facial Recognition Attendance System

Abstract

In this digital era, face recognition system plays a very important role in nearly every sector. Face recognition is one of the mostly used natural science. it'll used for security, authentication, identification, and has got a lot of blessings. Despite of obtaining low accuracy once compared to iris recognition and fingerprint recognition, it is being wide used due to its contactless and non-invasive technique. what's a lot of, face recognition system can even be used for attending marking in colleges, colleges, offices, etc. This system aims to make a class attending system that uses the thought of face recognition as existing manual attending system is time overwhelming and cumbersome to stay up. And there's conjointly prospects of proxy attending. Thus, the requirement for this technique can increase. this technique consists of four phases- data creation, face detection, face recognition, attending updating. Data is created by the pictures of the students in class. Face detection and recognition is performed exploitation Haar-Cascade classifier and native Binary Pattern chart algorithmic program severally. Faces unit detected and recognized from live streaming video of the room. attending are armored to the individual faculty at the tip of the session. it's standard that marking attending of the scholars is associate degree obligatory half in academe. standard technique of marking the attending is being followed by numerous establishments and Universities with several manual interventions. to scale back time consumption and human effort, the employment of associate degree automatic method of marking attending supported image process may be implemented. Authors have projected a sensible attending observance system through face detection and recognition techniques supported their face expression. a group of pictures of the scholars are antecedently fed to the system against that the live pictures of the scholars are compared and attending would be recorded supported facial characteristics. The projected approach uses CNN rule for coaching the pictures and LBPH visual descriptor for image classification. This models are going to be capable of providing higher degree of accuracy compared to already existing literature work. Authors have compared their experimental results with the present approaches and located satisfactory