Visual Recognized Attendance System

Abstract

Usually in our regular class we face a situation where faculty or respective administrators take attendance which would consume time. So, this project aims to develop a visual recognized attendance register that will identify the people based on the video and mark as present in the register. The data set is created by taking pictures of the people that are in a particular class. Based on the pictures the model will develop a identification mark or some important patterns for each student, based on this when we start a webcam through our model it will identify people and post attendance. We have Local Binary pattern histogram and convolutional neural network to identify each person uniquely. And to develop this we use OpenCV library which contains which contains many features regarding face recognition.