IMPLEMENTASI METODE TRANSFORMASI WAVELET DAN ALGORITMA K-NEAREST NEIGHBOR UNTUK PENGENALAN POLA MOTIF ULOS BATAK TOBA
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Silitonga, Samuel
Kuswardani, Dwina
Siregar, Riki Ruli A
Kuswardani, Dwina
Siregar, Riki Ruli A
Subject
Teknik Informatika
Datestamp
2022-09-22 06:08:36
Abstract :
Ulos is a woven fabric which is a cultural symbol of the Batak tribe which consists of various types of motifs. One method that can be used to identify ulos batak patterns based on their motives is the wavelet transformation method. The pattern recognition process begins by decomposing the ulos image which has been converted into a grayscale image with the Haar type wavelet transformation method then feature extraction with GLCM then using the k-nearest neighbor algorithm for the classification process. Test data used in the form of ulos images measuring 512x512 pixels jpg format as much as 25 data. Based on the results of the experiments that have been carried out, the highest level of recognition of the tested image is shown when the test data used by B1-B5 (Ulos Ragihotang) is classified with wavelet extraction data of each subband LL, LH, HL and HH by 70% and the lowest level of recognition when the test data used D1-D5 (Ulos Sibolang) is classified with wavelet extraction data for each subband LL, LH, HL, and HH by 25%.
Ulos is a woven fabric which is a cultural symbol of the Batak tribe which consists of various types of motifs. One method that can be used to identify ulos batak patterns based on their motives is the wavelet transformation method. The pattern recognition process begins by decomposing the ulos image which has been converted into a grayscale image with the Haar type wavelet transformation method then feature extraction with GLCM then using the k-nearest neighbor algorithm for the classification process. Test data used in the form of ulos images measuring 512x512 pixels jpg format as much as 25 data. Based on the results of the experiments that have been carried out, the highest level of recognition of the tested image is shown when the test data used by B1-B5 (Ulos Ragihotang) is classified with wavelet extraction data of each subband LL, LH, HL and HH by 70% and the lowest level of recognition when the test data used D1-D5 (Ulos Sibolang) is classified with wavelet extraction data for each subband LL, LH, HL, and HH by 25%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
