KLASIFIKASI JENIS KELAMIN MANUSIA MENGGUNAKAN CITRA WAJAH BERBASIS METODE SUPPORT VECTOR MACHINE (SVM); HUMAN GENDER CLASSIFICATION WITH FACIAL IMAGE BASED ON SUPPORT VECTOR MACHINE (SVM) METHOD

LATHIEF, DJAENUDDIN, Bambang Purwono

2016 | Skripsi | FMIPA

Face is a biometric features of a person. This biometric features encourages researchers to modeled it. This unique features modeled and apllied in many object. One of them is used to classifified the gender of someone like done in this research. First we must extract the faces. This extraction used to obtain an unique information named features. The method to get the features named Histogram of Gradient (HOG). After set the parameters in extracton method we will get many features named HOG features. Meanwhile, Support Vector Machine (SVM) used to classifiy the human gender. SVM was chosen because it’s one of classification method that robust in classifiy two group of data. The type of the kernel that used in this research is linear kernel. This kernel has advantage in the robustness to classify the big features. Implementation result show that the best achieved when parameters of cell size is (20x20 pixels), block size is (5x5 pixels), block stride is (5x5 pixels), bin = 5/7 and SVM parameters (C=1, epsilon = 1e-6 dan iterasi = 10.000). The accuration of that parameters is 92%.

Kata Kunci : Face, HOG Features, Linear SVM