Klasifikasi Lukisan Karya Van Gogh Menggunakan Convolutional Neural Network-Support Vector Machine

Abstract

Painting is a work of art with various strokes, textures, and color gradations so that a painting that is synonymous with beauty is created. The various paintings created have characteristics, such as the paintings by Van Gogh, which have tightly arranged strokes, creating a repetitive and patterned impression. This study classifies paintings by Van Gogh or not by using the VGG-19 and ResNet-50 feature extraction methods. The SVM method is used as a classification method with two optimizations, namely random and grid optimization in the linear kernel. The data set used consisted of 124 Van Gogh paintings and 207 paintings by other painters. The use of VGG-19 feature extraction using grid optimization has the best value of 93,28% using the use of random optimization which has a value of 92,89%. The use of ResNet-50 using grid optimization with the best value of 90,28% using the use of random optimization which has a value of 90,15%. The extraction feature of VGG-19 is better than ResNet-50 in paintings by Van Gogh or not.

Klasifikasi Kinship Keluarga Hollywood Dengan Metode Visual Geometry Group-Support Vector Machine

“…Metode CNN dapat dimodifikasi menjadi metode CNN-SVM dengan metode SVM [3] [14] [15]. Metode CNN-SVM dilakukan dengan cara mengenali wajah menggunakan metode CNN, kemudian hasilnya dimasukkan ke dalam metode SVM untuk mendapatkan model CNN-SVM.…”

Water quality identification based on remote sensing image in industrial waste disposal using convolutional neural networks

“…Cracks in masonry structures resemble grout lines, which can be mistaken for cracks. [9] stated that the CNN-SVM hybrid method could classify Van Gogh's paintings or not. The technique used is a patch image to divide each painting image into smaller parts and resize the image to a size of 224 x 224 pixels.…”