<![CDATA[The development of information technology is accompanied by advances in the fieldArtificial Intelligence, one of which is the facial recognition systemmachine learning ordeep learning. When humans see an image or video, we can recognize and find objects of interest in an instant. The goal of object detection is to replicate this intelligence using computers. The student attendance system at Bina Insan Lubuklinggau University is still done manually so it has many deficiencies such as attendance data or signatures that can be manipulated by others, causing fraud. there is a risk of losing data, in the process of inputting data. So that in this study a Student Attendance System was created at the Faculty of Engineering, Bina Insan University Using the YOLO v5 Algorithm. It is hoped that this will be used to obtain appropriate attendance results and there will be no more fraud and the risk of inputting attendance data errors, especially at Bina Insan University, Lubuklinggau City. In this study, 5 class datasets were used in the form of 5 students at the Faculty of Engineering, Bina Insan University. The accuracy results in detecting faces are approximately 80%. From a dataset of approximately 1500 images, with a photo size of 640x640pixel, 16 batch withepoch as many as 100 models will be evaluated by taking into account the values of mAP, Precision, Recall. In this study, the model gets 99% mAP, 99% precision, and 99% recall, with this, the system created is good enough]]>
Copyrights © 2023
