Automatic Attendance System for Using Face Recognition Based on Deep Learning

Abstract

Student participation is fundamental in the learning procedure. To record participation, a few different ways should be possible; one of them is through student marks. The procedure has a few weaknesses, for example, requiring quite a while to make attendance; the attendance paper is lost; the organization must enter participation information individually into the PC. In the existing technology or method(LBPH). It is having lot of demerits like it is difficult for the computer to do the face identification when the poses of the probe are different. Due to other objects or accessories (e.g., sunglasses, scarf, etc.) performance of face recognition algorithms gets affected. To beat this, the paper proposed an understudy participation framework that utilizations face acknowledgment. In the proposed framework, Convolutional Neural Network (CNN) is utilized to distinguish faces in images, profound measurement learning is utilized to create facial implanting, and K-NN is utilized to group understudy's countenances. In this manner, the PC can perceive faces. From the tests led, the framework had the option to perceive the essences of understudies who did join in and their participation information was naturally spared. For proposed CNN we have obtained a best recognition accuracy of 98.3 %. The proposed method based on CNN outperforms the state of the art methods.