Sistem Pendeteksi Kulit Manusia Menggunakan Segmentasi Warna Kulit Pada Tipe Citra HSV (Hue Saturation Value)

https://doi.org/10.22146/ijeis.2434

Ilona Usuman(1*), Andi Dharmawan(2), Aufaclav Zatu Kusuma Frisky(3),

(1) 
(2) 
(3) 
(*) Corresponding Author

Abstract


Abstrak

Telah dibuat suatu sistem pendeteksi kulit manusia melalui citra HSV ( Hue Saturation Value ) menggunakan program Matlab. Penelitian ini dilakukan karena masih minimnya penelitian untuk citra HSV pada Matlab. Berbedanya nilai Hue dan Saturation pada program matlab maka penelitian ini akan berguna untuk pendeteksian kulit m anusia menggunakan program Matlab.

            Penelitian ini dimulai dengan mencari nilai Hue dan Saturation pada citra HSV dan menentukan range nilai kulit manusia yang dapat dibedakan juga dengan objek yang memiliki warna mirip dengan kulit manusia. Hasil dari segmentasi kemudian di filter menggunakan low pass filter untuk menghilangkan noise. Hasilnya akan dideteksi untuk membedakan kulit manusia yang berupa muka dan selain muka.

Penelitian ini menghasilkan kesimpulan range kulit manusia memiliki nilai Hue sebesar 0.021 sampai 0.080 dan Nilai Saturation sebesar 0.23 sampai 0.68. Metode ini dapat bekerja dengan sangat baik dengan kecepatan rata-rata 1.4 frame per second dengan tingkat akurasi 87.86 %. Akurasi keberhasilan dipengaruhi oleh gerakan objek, posisi dan arah objek wajah terhadap kamera, jumlah objek orang dalam satu frame, jarak objek kulit manusia dengan kamera serta intensitas cahaya pencahayaan.

 

Kata Kunci : Skin, Hue, Saturation, HSV, segmentasi

 

Abstract

System for human skin detection through image HSV (Hue Saturation Value) using Matlab program has been made. The research was carried out because it is still lack of research for HSV image on Matlab. Different value of Hue and Saturation in Matlab program become the main reason of this research,  and this research will be useful for the detection of human skin using Matlab program for the future.

The research began with a search for the value of Hue and Saturation in HSV image and determine the range of values that the human skin can be distinguished also by objects that have a color similar to human skin. The results of segmentation then filtered using a low pass filter to eliminate noise. The result will be detected to distinguish human skin in the form of face to face and in addition to giving a red box for the front and green for other than the face.

This research resulted in the conclusion of human skin has a value range of 0.021 to 0.080 Hue and Saturation Value for 0.23 until 0.68. This method can work very well with an average speed of 1.4 frames per second with an accuracy rate of 87.86%. Accuracy is influenced by the success of object movement, position and direction of the object's face to the camera, the number of objects in a single frame, the distance of the object to the camera and the human skin light intensity lighting.

 

Keywords:  Skin, Hue, Saturation, HSV, segmentasi


Full Text:

PDF



DOI: https://doi.org/10.22146/ijeis.2434

Article Metrics

Abstract views : 1055 | views : 526

Refbacks

  • There are currently no refbacks.




Copyright (c) 2017 IJEIS - Indonesian Journal of Electronics and Instrumentation Systems

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



Indonesian Journal of Electronics and Instrumentations Systems
(IJEIS) ISSN 2088-3714 (print); ISSN 2460-7681 (online)
is a scientific journal the results of Electronics
and Instrumentations Systems
A publication of IndoCEISS.
Gedung S1 Ruang 416 FMIPA UGM, Sekip Utara, Yogyakarta 55281
Fax: +62274 555133
email:ijeis.mipa@ugm.ac.id | http://jurnal.ugm.ac.id/ijeis


Creative Commons License
IJEIS by http://jurnal.ugm.ac.id/ijeis is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
View My Stats1
View My Stats2