PENGENALAN EKSPRESI WAJAH MENGGUNAKAN METODE LOCAL BINARY PATTERN HISTOGRAM DAN JARINGAN SYARAF TIRUAN
Daftar Isi:
- Facial expressions are basically non-verbal communication for humans to convey emotions and intentions to others during interactions. This study aims to recognize each facial expression that will be displayed whether it is suitable or not. In conducting the introduction of a facial expression of required some the process in which the image of the face can be recognized correctly, it is to do detection in the image of using haar classifier on library opencv. Next, face extraction is performed to capture features using the Local Binary Pattern Histogram method, where each segment of the face image will be divided into smaller areas with a certain number of blocks according to the pixel size of the data. For the process of recognizing facial expressions using the Backpropagation Artificial Neural Network method where the data will be trained and tested so that the expression results can be known. The results of this study indicate that each data that has been trained and reaches the specified error value will get a higher accuracy of 90%, while the data tested without training will get a much smaller accuracy.