PERBANDINGAN METODE K-NEAREST NEIGHBORS DAN RANDOM FOREST UNTUK KLASIFIKASI WAJAH PADA CUPLIKAN VIDEO DENGAN PENDEKATAN LOCAL BINARY PATTERN DAN HISTOGRAM WARNA (HUE, SATURATION, VALUE) (STUDI KASUS: LIMA ANGGOTA TOMORROW X TOGETHER DARI EMPAT MUSIK V

Gicela Viga Gita Wisesa, NIM.: 22106010023 (2026) PERBANDINGAN METODE K-NEAREST NEIGHBORS DAN RANDOM FOREST UNTUK KLASIFIKASI WAJAH PADA CUPLIKAN VIDEO DENGAN PENDEKATAN LOCAL BINARY PATTERN DAN HISTOGRAM WARNA (HUE, SATURATION, VALUE) (STUDI KASUS: LIMA ANGGOTA TOMORROW X TOGETHER DARI EMPAT MUSIK V. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

	Text (PERBANDINGAN METODE K-NEAREST NEIGHBORS DAN RANDOM FOREST UNTUK KLASIFIKASI WAJAH PADA CUPLIKAN VIDEO DENGAN PENDEKATAN LOCAL BINARY PATTERN DAN HISTOGRAM WARNA (HUE, SATURATION, VALUE) (STUDI KASUS: LIMA ANGGOTA TOMORROW X TOGETHER DARI EMPAT MUSIK V) 22106010023_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (6MB)
	Text (PERBANDINGAN METODE K-NEAREST NEIGHBORS DAN RANDOM FOREST UNTUK KLASIFIKASI WAJAH PADA CUPLIKAN VIDEO DENGAN PENDEKATAN LOCAL BINARY PATTERN DAN HISTOGRAM WARNA (HUE, SATURATION, VALUE) (STUDI KASUS: LIMA ANGGOTA TOMORROW X TOGETHER DARI EMPAT MUSIK V) 22106010023_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (14MB) | Request a copy

Abstract

Face classification is the process of grouping individual identities based on visual facial characteristics represented in the form of numeric features through digital image processing and machine learning. Human faces have distinctive texture and color distribution patterns that can be used as distinguishing features between individuals. This study discusses face classification in video excerpts using texture and color features, namely the Local Binary Pattern and Color Histogram methods in the HSV color space. These visual features are then used as the basis for classification using the K-Nearest Neighbors (KNN) and Random Forest methods. Evaluation using a confusion matrix and measurements of accuracy, precision, recall, and F1-Score showed that the combination of LBP and HSV features was effective in representing facial characteristics in video footage. The test results show that KNN produces an accuracy of 60% with a precision of 0.47, a recall of 0.50, and an F1-score of 0.60, while Random Forest shows better performance with an accuracy of 80% with a precision of 0.70, a recall of 0.80, and an F1-score of 0.73, so it can be concluded that the combination of HSV and LBP features with the Random Forest algorithm is more effective for face classification in K-POP music video clips in the case study conducted.

Share this knowledge with your friends :

Actions (login required)

View Item