Face Recognition using Modified Triangle Method

  • Satriya A
  • Agustini S
N/ACitations
Citations of this article
20Readers
Mendeley users who have this article in their library.

Abstract

Abstrak. Pengenalan wajah diperlukan untuk beberapa aplikasi seperti siatem keamanan. Pada penelitian ini, peneliti memberikan metode triangle yang baru untuk deteksi dan pengenalan wajah dari wajah seseorang yang diaplikasikan untuk sistem keamanan. Metode ini berdasarkan deteksi warna kulit untuk mendeteksi wajah manusia dan mengenali wajah tersebut menggunakan metode modified triangle. Yang dilakukan pertama kali adalah pemrosesan gambar dimana semua gambar dijadikan dalam resolusi yang sama. Kemudian, metode ini akan melakukan skin detection berdasarkan warna kulit dari gambar, non-face object akan dihapus oleh morphological method untuk mendapatkan hanya gambar wajah saja, segmentasi wajah, deteksi features point dari wajah (mata kanan, mata kiri, dan mulut) dan kemudia dihitung Euclidean distance diantara features face. Jarak antara eatures face akan dibandingkan dengan database untuk proses pengenalan wajah. Dari hasil penelitian menunjukkan bahwa metode ini dapat mencapai tingkat keberhasilan 90% dalam deteksi wajah dan 50% untuk pengenalan wajah. Kata kunci: deteksi wajah, pengenalan wajah, triangle method Abstract. Human face recognition is required for several applications such as security system. In this paper, we propose a new triangle method for detection and recognition of the human face which is applied for security system. We use the skin colour based detection to detect the human face and recognize the selected face by modified triangle method. First, we do image pre-processing which include change all picture in same resolution. Next, this method will do skin detection based skin colour from the image, nonface object removal by morphological method for getting the face of image only, face segmentation, detect features point of face (right eye, left eye, and mouth) and then compute Euclidean distance between features face. The distance between features face will be compared with database for face recognition. Experimental result shows that our method can achieve 90% success rate for face detection and 50% for face recognition. Keywords : face detection, face recognition, triangle method

Cite

CITATION STYLE

APA

Satriya, A. B., & Agustini, S. (2017). Face Recognition using Modified Triangle Method. INTEGER: Journal of Information Technology, 2(1). https://doi.org/10.31284/j.integer.2017.v2i1.92

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free