Cara Cepat Untuk Mendeteksi Keberadaan Wajah Pada Citra Yang Mempunyai Background Kompleks Menggunakan Model Warna YCbCr dan HSV

Penulis

  • Nurul Hidayat
  • Muh. Arif Rahman

DOI:

https://doi.org/10.25126/jtiik.201522147

Abstrak

Abstrak

Pendeteksi wajah dari sebuah citra baik berupa gambar diam maupun bergerak merupakan topik penting dan menarik saat ini. Proses deteksi keberadaan wajah ini menjadi dasar dari proses pengenalan wajah yang mempunyai banyak implementasi baik pada bidang keamanan maupun sosial media. Tujuan dari proses deteksi wajah adalah untuk mengetahui apakah ada wajah dari suatu citra, kemudian menemukan letak keberadaan wajah. Pendeteksian wajah adalah tahapan penting dari aplikasi yang memanfaatkan keberadaan wajah pada suatu citra. Implementasinya cukup banyak terutama di bidang biometri keamanan dan sosial media. Riset ini mengusulkan deteksi wajah menggunakan 3 tahapan umum yaitu segmentasi warna kulit manusia, binarisasi dan penentuan region garis serta deteksi wajah menggunakan ruang warna YCbCr dan HSV. Dalam penelitian ini dilakukan deteksi wajah pada 10 citra yang memiliki background yang kompleks. Pendeksian lokasi wajah didasarkan pada temuan hole mata yang simetris. Wajah yang terlalu kecil membuat keberadaan mata hanya terdeteksi sebelah sehingga mengakibatkan wajah tidak terdeteksi. Hasil evaluasi didapatkan tingkat akurasi rata-rata deteksi wajah mencapai 83,4% dengan kecepatan rata-rata 6530 piksel/detik.

Kata kunci: Deteksi Wajah, Biometri, Segmentasi, YCbCr, HSV, Region Garis

Abstract

Face detection of an image either still or moving image is an important and interesting topic today. Face detection process where it became the basis of face recognition process that has many implementations, both in the field of security and social media. The aim of the face detection process is to determine whether there is a face from an image, and then locate the whereabouts of the face. Face detection of an image, either a statis or moving image  Face detection is an important phase in application system to utilize face location in an image. It is so many implementations such asspecially for security and sociality biometrics field. Here, it suggest to detect the face location with 3 steps, skin human color segmentation step, binnerization step, and locate line region then detection step of face location by YCbCr and HSV color region. Here, ten images which have a complex background are implemented. To detect face location based on finding symetris of eyes hole. The faces which are too small made the eyes locations are only detected a half, so the face can’t detect. The evaluation result obtained average of accuration of face detection 83,4% with average time to detect 6530 pixel/second.

Keywords: Face detection, Biometry, Segmentation, YCbCr, HSV, Line Region

Downloads

Download data is not yet available.

Referensi

Chen Aiping, Pan Lian, Tong Yaobin, Ning Ning, 2010. Face Detection Technology Based on Skin Color Segmentation and Template Matching. Educ. Technol. Comput. Sci. ETCS 2010 Second Int. Workshop On 2, 708–711.

Ghimire, Deepak ;, Lee, Joonwhoan, 2013. A Robust Face Detection Method Based on Skin Color and Edges. J. Inf. Process. Syst. 9, 141–156.

Goel, P., Agarwal, S., 2012. An Illumination Invariant Robust and Fast Face Detection, Feature Extraction Based Face Recognition System. Comput. Commun. Technol. ICCCT 2012 Third Int. Conf. On 110–115.

Han, J., Awad, G., Sutherland, A., 2009. Automatic skin segmentation and tracking in sign language recognition. Comput. Vis. IET 3, 24–35.

Kouzani, A.Z., He, F., Sammut, K., 1997. Commonsense knowledge-based face detection. Intell. Eng. Syst. 1997 INES 97 Proc. 1997 IEEE Int. Conf. On 215–220.

Lijing Zhang, Yingli Liang, 2010. A fast method of face detection in video images. Adv. Comput. Control ICACC 2010 2nd Int. Conf. On 4, 490–494.

Muh. Arif Rahman, 2014. Deteksi Warna Kulit Manusia Berbasis Piksel Menggunakan Ruang Warna HSV Dan YCbCr. Presented at the KNSI 2014, Makasar Indonesia.

Pan Ng, Chi-Man Pun, 2011. Skin Color Segmentation by Texture Feature Extraction and K-mean Clustering. Comput. Intell. Commun. Syst. Netw. CICSyN 2011 Third Int. Conf. On 213–218.

Rita, V., Anupam Agrawal, Shanu Sharma, 2010. A Robust & Fast Face Detection System. ACEEE Int J Signal Image Process., 03 01, 17–22.

Sigal, L., Sclaroff, S., Athitsos, V., 2000. Estimation and prediction of evolving color distributions for skin segmentation under varying illumination. Comput. Vis. Pattern Recognit. 2000 Proc. IEEE Conf. On 2, 152–159 vol.2.

Son Lam Phung, Chai, D., Bouzerdoum, A., 2003. Adaptive skin segmentation in color images. Multimed. Expo 2003 ICME 03 Proc. 2003 Int. Conf. On 3, III–173–6 vol.3.

Tian Qing, Zhao Shiwei, 2012. A fast face detection method for JPEG image. Signal Process. ICSP 2012 IEEE 11th Int. Conf. On 2, 899–902.

Wang, J., Yang, H., 2008. Face Detection Based on Template Matching and 2DPCA Algorithm. Image Signal Process. 2008 CISP 08 Congr. On 4, 575–579.

Wei Wang, Yongsheng Gao, Siu Chueng Hui, Karhang Leung, 2002. A fast and robust algorithm for face detection and localization. Neural Inf. Process. 2002 ICONIP 02 Proc. 9th Int. Conf. On 4, 2118–2121 vol.4.

Yuanyuan Liu, Haibin Yu, Zhiwei He, Xueyi Ye, 2009. Fast Robust Face Detection under a Skin Color Model with Geometry Constraints. Comput. Intell. Secur. 2009 CIS 09 Int. Conf. On 2, 515–519.

Zhanwu Xu, Miaoliang Zhu, 0-0 0. Color-based skin detection: survey and evaluation. Multi-Media Model. Conf. Proc. 2006 12th Int. 10 pp.

Unduhan

Diterbitkan

22-07-2015

Terbitan

Bagian

Teknologi Informasi

Cara Mengutip

Cara Cepat Untuk Mendeteksi Keberadaan Wajah Pada Citra Yang Mempunyai Background Kompleks Menggunakan Model Warna YCbCr dan HSV. (2015). Jurnal Teknologi Informasi Dan Ilmu Komputer, 2(2), 138-142. https://doi.org/10.25126/jtiik.201522147