Pengenalan Motif Songket Palembang Menggunakan Deteksi Tepi Canny, PCA dan KNN

Abstract

Palembang Songket is one type of songket characteristic of Indonesian culture which has various types of motifs. Various types of motifs make it difficult for ordinary people to recognize songket that has a similar motif. This study aims to identify 2 types of Palembang songket motifs, namely bintang berante and nampan perak. The classification process will go through 3 stages, namely preprocessing, feature extraction and classification. The preprocess changes the color image of songket into grayscale image. In the feature extraction stage, the grayscale image is increased in contrast with histogram equalization and then uses Canny edge detection to obtain patterns from the songket motif. The extraction results are then grouped and labeled according to their motives for further classification using Principal Component Analysis (PCA) and k-Nearest Neighbor (KNN). Previous studies have resulted in an accuracy of 90.5% for 5 types of motifs using the back propagation algorithm. From the trial results obtained an accuracy value reached 91.67%.

Classification of Pringgasela Typical Songket Using Multi Texton Co-occurrence Descriptor and K-Nearest Neighbor

“…Similarly, research introducing Balinese traditional Songket cloth using the KNN method resulted in an accuracy of 77.3% [6]. Other research that supports the classification process using the KNN method is the recognition of Palembang Songket Motifs Using the Canny Edge Detection method, PCA and KNN gave 91.67% accuracy using 52 typical Palembang Songket data [5]. Furthermore, research on the implementation of the K-Nearest Neighbor (KNN) Classification Method for Lampung Batik Pattern Recognition has been successfully carried out, where the level of accuracy obtained is 98.12% [12].…”

“…In selecting a class at KNN, it is done by looking for group data with the closest distance (k) from the training data object that is closest to the new data that does not yet have that class. The KNN method has been used to identify Palembang Songket types with the highest accuracy of 91.67% [5]. There are also other experiments to carry out the Songket recognition process using the KNN method, namely the introduction of Balinese Songket cloth which in this study resulted in an accuracy of 77.3% [6].…”

Deteksi Sarung Samarinda Menggunakan Metode Naive Bayes Berbasis Pengolahan Citra

“…Pada pre-processing diimplemetasikan image resizing [7], grayscaling [8], [9], dan histogram equalization [8]. Berikutnya, ekstraksi fitur dilakukan berdasarkan warna, bentuk, atau tekstur.…”

“…Fitur warna dapat dihasilkan menggunakan metode color moments [9], color histogram [7]. Sementara itu, fitur bentuk menggunakan moment invariant [10], deteksi tepi dengan operator Canny [8], [11], Sobel [12], [13], Prewitt dan Gaussian [14], sedangkan fitur tekstur menggunakan GLCM [15]- [20], Local Binary Pattern (LBP) [21], HMTSeg [22], dan Filter Gabor [3], [23]. Pada proses akhir yaitu klasifikasi, metode yang umum digunakan dalam pengenalan kain tradisional adalah K-nearest neighbor (KNN) [8], [9], [18], [20], [24], [25], probability neural network (PNN) [3], jaringan syaraf tiruan (JST) [16], [26], support vector machine (SVM) [5], Naive Bayes [7], dan learning vector quantization [27].…”

Sistem Cerdas dalam Mengidentifikasi Kematangan Buah Naga Berdasarkan Fitur Tekstur dengan Metode K-Nearest Neighbor

“…Moh. Arie Hasan dengan menggunakan algoritma Principal Component Analysis dan KNearest Neighbor dapat digunakan dalam proses klasifikasi jenis songket [9]. Penelitian atas nama Feri Agustina, Dalam penelitian ini di peroleh tigkat keakuratan system ini yaitu 85% [10] Penelitian atas nama Chandra Wijaya, Hasil penelitian menunjukkan bahwa akurasi terbaik per kelas adalah 66,20% untuk K = 5 [11].…”