Smart Attendance System Using Machine Learning Algorithms

In face recognition, the difficult task is to detect the face from the existing database, and the problems may arise due to bad image quality, bad lighting, or the person not facing the camera, etc. The face of the same person will change for different expressions. If the person's age changes, the size, and color of the person may also change, which makes face recognition more difficult. In this paper, the proposed system android application is used for face recognition for attendance system. The proposed system can be used in colleges and schools for attendance systems. The human face will be recognized from the image uploaded to the server of that particular class; this can be done by using an algorithm. Here we are using the HAAR cascade classifier to identify the face by removing noise and crop the face in the image. Then, at that point the HOG (Histogram Of Gradients) calculation is utilized to extricate the features from the detected face. In the last step, we use SVM (Support Vector Machine) classifier to identify the person from our database. When the person is found in the image, then the attendance will be automatically allocated in the excel sheet of that particular class.