Songket fabric is a cultural heritage of Indonesia woven with gold or silver threads, creating textiles that are not only visually captivating but also rich in cultural significance. Each motif on Palembang Songket reflects the traditions and beliefs of the community, where the selection of motifs is often tailored to specific event contexts. However, the recognition of several motifs with similar patterns presents unique challenges in the identification process. This study aims to implement a Convolutional Neural Network (CNN) method for classifying Palembang Songket motifs. The dataset used consists of images of Songket motifs, including Bintang Berantai, Naga Besaung, Nampan Perak, and Pulir. The ResNet-50 architecture is utilized as the classification model. The results indicate that the implemented model achieves an accuracy of 96% in recognizing these motifs, thereby contributing to the preservation and enhancement of understanding regarding the cultural richness of Palembang Songket.
                        
                        
                        
                        
                            
                                Copyrights © 2024