Mendeteksi Ekspresi Wajah dengan Meninjau Iris Mata Menggunakan Metode Transformasi Hough dan K-Nearest Neighbor (KNN)

Ningsih Djamsi (Universitas Muhammadiyah Malang)
Didih Rizki Chandranegara (Universitas Muhammadiyah Malang)
Zamah Sari (Universitas Muhammadiyah Malang)

Jurnal Repositor
Vol 5 No 1 (2023): Februari 2023

The face has differences (characteristics), besides that the face is also the center of expression in humans. Directly facial expressions become a non-verbal communication, many factors can support detecting a person's facial expressions such as movements and changes in eyebrows, lips and eyes. In this study, the focus is on the iris as the object of observation. There are two test scenarios to detect facial expressions based on facial images, using the TFEIDHigh dataset. The first scenario uses the KNearest Neighbor (K-NN) method with an accuracy quality of 80.0% and an execution time of 61.3831 Seconds. The second scenario uses the Convolutional Neural Network (CNN) method with an accuracy of 80.0% with an execution time of 656,730 Seconds.