Identifikasi Landmark Ikonik Menggunakan Metode Histogram Of Oriented Gradient (hog) Dan Support Vector Machine (svm)

Abstract

Abstract Identifikasi landmark oleh sistem sangat sulit dilakukan karena untuk mengidentifikasi sistem harus mengenali ciri – ciri data dari citra tersebut, agar sistem dapat mengenali ciri - ciri citra landmark maka citra tersebut harus di feature extraction dan dilakukan pembalajaran yang dijadikan kecerdasan buatan untuk komputer. Indentifikasi Landmark Ikonik Menggunakan Metode Histogram of Oriented Gradient (HOG) dan Support Vector Machine (SVM) merupakan sebuah sistem yang dapat mengidentifikasi objek landmark yang ada pada citra dengan mengambil lapisan-lapisan seluruh citra gambar. Lapisan-lapisan citra di feature extraction menggunakan metode Histogram of Oriented Gradient (HOG) kemudian lapisan-lapisan citra yang di feature extraction dengan data training di klasifikasikan menggunakan Support Vector Machine (SVM) sehingga sistem dapat mencari posisi dan mengidentifikasi nama landmark apa yang ada pada citra. Dari hasil pengujian yang telah dilakukan mendapati bahwa identifikasi landmark ikonik menggunakan metode Histogram of Oriented Gradient dan Support Vector Machine mendapatkan akurasi sebesar 84,4% kebenaran. Kata kunci : support vector machine, histogram of oriented gradient, landmark, ikonik Abstract: Identification of landmarks by the system is very difficult to do because to identify the system, it must recognize the data features of the image, so that the system can recognize the features of the landmark image, the image must be feature extraction and a lesson made into artificial intelligence for the computer. Identification of Iconic Landmarks Using the Histogram of Oriented Gradient (HOG) Method and Support Vector Machine (SVM) is a system that can identify landmark objects in an image by taking the layers of the entire image image. The image layers in feature extraction use the Histogram of Oriented Gradient (HOG) method then the image layers that are feature extraction with training data are classified using a Support Vector Machine (SVM) so that the system can search for positions and identify what landmark names are on imagery. From the results of the tests that have been done, it is found that the identification of iconic landmark using the Histogram of Oriented Gradient and Support Vector Machine method gets an accuracy of 84.4%. Keywords: support vector machine, histogram of oriented gradient, landmark,iconic