Controlling educational attending situations by face detection by deep learning methods

With the developing technology, computers have become more powerful in terms of hardware and the increase in the processing capacity of processors has facilitated the processing of images taken instantly and in real time. Face recognition is one of the studies in the field of image processing. Face recognition is frequently used in security and commercial applications. The high achievements of artificial intelligence (AI) studies, especially in the last 20 years, have contributed to the spread of these studies to many different fields. Education is one of them. The potentials and advantages of using AI in education can be categorized under three headings: students, teachers and institutions. One of the institutional studies can be the security of educational environments and automation contribution to educational processes. Based on this, deep learning methods, one of the sub-branches of the field of AI, were used in this study. Since deep learning algorithms require large amounts of data and time, transfer learning is recommended. Transfer learning is the process of transferring the high performance deep learning method to the solution of the designer's problem. In this thesis, the dlib library, which is useful on low-resource devices, is used for transfer learning. Using image processing algorithms for object detection from images, a pioneering work has been designed and successfully implemented to keep the entrance records of students to the educational institution and to perform class attendance with the images taken from the camera.