Segmentasi Citra Wajah dengan Metode Learning Vector Quantization pada Gambar Resolusi Rendah
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Febianti, Alya
Haris, Abdul
Elly, Muhamad Jafar
Haris, Abdul
Elly, Muhamad Jafar
Subject
Teknik Informatika
Datestamp
2022-09-26 08:02:15
Abstract :
Human faces have similarities with each other so that it will be difficult to detect. In this study, the aim of this research is to make it easier to detect faces from the results of the webcam camera images whose resolution is still low. To solve the problem in this study, using the Learning Vector Quantization method for classifying facial images. In this study, the first time an image acquisition was carried out on a webcam camera, then the RGB value was taken and converted into greyscale form and then converted into binner by way of canny edge detection, the binary results were used as the input value in the calculation stage of the Learning Vector Quantization value. With this method, it is more appropriate to use it directly in the case of face detection because this method focuses on the citra classification and the accuracy results obtained are 88% with a comparison of training data and test data.
Human faces have similarities with each other so that it will be difficult to detect. In this study, the aim of this research is to make it easier to detect faces from the results of the webcam camera images whose resolution is still low. To solve the problem in this study, using the Learning Vector Quantization method for classifying facial images. In this study, the first time an image acquisition was carried out on a webcam camera, then the RGB value was taken and converted into greyscale form and then converted into binner by way of canny edge detection, the binary results were used as the input value in the calculation stage of the Learning Vector Quantization value. With this method, it is more appropriate to use it directly in the case of face detection because this method focuses on the citra classification and the accuracy results obtained are 88% with a comparison of training data and test data.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
