DETEKSI JERAWAT PADA WAJAH MENGGUNAKAN METODE VIOLA JONES

Haruno Sajati, Yuliani Indrianingsih, Puspa Ira Dewi Candra Wulan

Abstract

Acne is a skin disease that most often grow in the face and very disruptive to affect the appearance of a person's face. Viola Jones is one method to help detect acne so someone get information about the state of the face and immediately overcome. Viola Jones Method has three main processes integral image is used to determine whether there is a feature haar particular in an image, the method adaboost machine learning is used to select features haar specific that will be used to adjust the threshold value, and a cascade classifier as the classification of the final determining regions the face in the picture. Testing of 30 samples, the results obtained showed that the method is less accurate Viola Jones is used as a method of detection of acne with results average 25 % the percentage of common acne, blackheads 0 % and 45% of cystic acne. The number of samples used to create xml greatly affect the results of detection.

Keywords

Acne, Viola Jones, Integral Image, Adaboost Machine Learning, Cascade Classifier.

References

Adi Siswando, dkk. 2013. Algoritma C4.5 Berbasis Adaboost untuk Prediksi Penyakit Jantung Koroner: Fakultas Teknik Komputer Universitas Sains Al-Quran Jawa Tengah Wonosobo.

Daniel Lelis Baggio. 2015. Open CV Computer Vision with Java. Birmingham, Mumbai.

Hadisantoso dan Agus Harjoko. 2013. Haar Cascade Classifier dan Algoritma Adaboost untuk Deteksi Banyak Wajah Dalam Ruang Kelas. Universitas Gadjah Mada Yogyakarta.

Harper,J.C. (2007). Acne Vulgaris. Birmington : Departement of Dermatology University of Alabama.

Kumar S., Prasad S., 2011. Real Time Face Recognition Using Adaboost Improved Fast PCA Algorithm, Department of Computer Engineering, Ideal Institute of Technology, Ghaziabad, INDIA.

Mahdi Rezai,2013. Creating a Cascade of Haar-Like Classifiers: Step by Step, Department of Computer Science, the University of Auckland.

Article Metrics

Abstract view: 833 times
Download     : 782   times

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Refbacks

  • There are currently no refbacks.