Penerapan Metode K-Nearest Neighbor untuk Klasifikasi Jenis Jerawat pada Wajah Menggunakan Fitur GLCM
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Amanda, Muhammad Raju
Kuswardani, Dwina
Djunaidi, Karina
Kuswardani, Dwina
Djunaidi, Karina
Subject
Teknik Informatika
Datestamp
2022-09-27 02:13:26
Abstract :
Acne is a lump that usually occurs on facial skin. Acne can also arise due to the pilosebaceous unit gland's inflammation due to keratin blockage in the skin. This study discusses the classification of acne types on the face using the Gray Level Co- Occurance Matrix (GLCM) feature and the K-Nearest Neighbor (KNN) method. The stages in the GLCM method use the parameters Contrast, Correlation, Energy, an Homogeneity and use the directions of 0 degrees, 45 degrees, 90 degrees, and 135 degrees. The purpose of this study was to classify the types of acne. To get the value of each parameter, we must use the grayscale image feature. There are four acne types, namely cystic acne, pustular acne, blackhead, and whitehead. The total image data used is 80 image data, of which 60 image data is used for training data, and 20 image data is used as test data. All images are 255 x 255 pixels in size to simplify the classification process. The classification results using the GLCM feature show an overall accuracy rate of 70%.
Acne is a lump that usually occurs on facial skin. Acne can also arise due to the pilosebaceous unit gland's inflammation due to keratin blockage in the skin. This study discusses the classification of acne types on the face using the Gray Level Co- Occurance Matrix (GLCM) feature and the K-Nearest Neighbor (KNN) method. The stages in the GLCM method use the parameters Contrast, Correlation, Energy, an Homogeneity and use the directions of 0 degrees, 45 degrees, 90 degrees, and 135 degrees. The purpose of this study was to classify the types of acne. To get the value of each parameter, we must use the grayscale image feature. There are four acne types, namely cystic acne, pustular acne, blackhead, and whitehead. The total image data used is 80 image data, of which 60 image data is used for training data, and 20 image data is used as test data. All images are 255 x 255 pixels in size to simplify the classification process. The classification results using the GLCM feature show an overall accuracy rate of 70%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
