The Artificial Neural Networks (ANN) for Batik Detection Based on Textural Features

Abstract

This study aims to utilize artificial neural networks to distinguish batik motifs and non-batik fabric motifs. Several important steps are needed, namely the process of acquiring batik and non-batik images, pre-transforming batik and non-batik images to gray scale forms, texture feature extraction in gray scale images and detection of motifs using networks artificial nerve. Image acquisition is done by collecting batik and not batik images from several different motifs. Processing data sets is divided into 70% as training data and 30% as testing data. Artificial neural network models used in this research use the Backpropagation learning algorithm by comparing the Scaled conjugate gradient algorithm (trainscg) training method and the Levenberg-Marquardt algorithm (trainlm) training method. The results obtained for the accuracy using the Scaled conjugate gradient algorithm (trainscg) training method were higher with an accuracy value of 84.12%, compared to the Levenberg-Marquardt algorithm (trainlm) method by 86.11%.

Deteksi Sarung Samarinda Menggunakan Metode Naive Bayes Berbasis Pengolahan Citra

“…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].…”

“…Model JST yang digunakan dalam penelitian ini adalah algoritma pembelajaran backpropagation dengan pelatihan scaled conjugate gradient (trainscg) dan metode pelatihan Lavenberg-Marquardt (trainlm). Hasil yang diperoleh untuk akurasi dengan metode pelatihan algotirma scaled conjugate gradient (trainscg) lebih tinggi dengan nilai akurasi sebesar 84.12%, dibandingkan dengan metode algoritma Lavenberg-Marquardt (trainlm) sebesar 86.11% [26].…”

Pattern Recognition of Sarong Fabric Using Machine Learning Approach Based on Computer Vision for Cultural Preservation

“…Indonesia has the traditional fabric recognized as one of the country's cultural heritage in the form of batik and sarong fabric. Batik has various patterns and motifs which derive from several areas, such as Java [10,11], central Sulawesi [9], and Madura [12]. Samarinda is one of the cities in East Kalimantan-Indonesia's largest provincehas a distinctive sarong that residents typically wear for formal or religious occasions.…”

“…Generally, those works involved three main processes: preprocessing, feature extraction, and classification [8,9,14]. Pre-processing was commonly performed by resizing the original image into a square form [8,9], converting the RGB image into grayscale [11], histogram equalization [15], edge detection [13], and Gaussian filter [3].…”

“…Three types of features were extracted throughout the feature extraction process: color [12,16,17], texture [10,11], and shape [10,12,16,17]. Color moments and histograms were used to generate the color features [12].…”