Employee Attendance Through Face Recognition Using the HAAR Cascade Classifier Method

Human resource management in an organization is a very important factor. The same is done by the management of Mbak Dwi's restaurant. However, until 2022, the management of employee attendance still uses a manual system. Manual attendance systems often cause attendance data to be at risk of being manipulated so that the data is invalid. The purpose of this study is to create a presence system with face detection using the HAAR Cascade Classifier method. The methods used in this study include the literature method, an observation method and a prototyping method. This research produced a presence system using face detection. The test results show that the system was able to detect and identify faces well with the use of 5W (261 Lux) LED lights at a distance of 20 to 40 centimeters. In dry face conditions, it has a very good accuracy value at a distance of 20 to 40 centimeters. Recording the employee's face was carried out at a distance of 20 to 40 centimeters using a light approximately 261 Lux with dry face condition. This is done so that the image of the face can be captured by the attendance system perfectly. There were several drawbacks to the use of the HAAR Method, namely the face position, the image shooting condition, the direction of the light source, and the characteristics of the sensor and camera lens.